(WRC-SouthAfrica-2001-R. McKenzei) PRESMAC User Guide

Pressure Management Program (PRESMAC) TT152/01
PRESMAC
Pressure Management Program
developed through
SOUTH AFRICAN WATER RESEARCH

COMMISSION
By
Ronnie McKenzie : WRP Pty Ltd.
(Coding by Stephen Langenhoven)
WRC Report No TT 152/01
March 2001
i 02/05/16
Obtainable from;
Water Research Commission

PO Box 824
PRETORIA
0001
The publication of this report emanates from a project entitled: Water Leakage:
Pressure Management Model (WRC Project Number K5/997)
DISCLAIMER
This report has been reviewed by the Water Research Commission (WRC) and
approved for publication. Approval does not signify that the contents necessarily
reflect the views and policies of the WRC, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
ISBN No. 1 86845 722 2
ii 02/05/16
IMPORTANT
PREFACE
This document incorporates the user guide to the South African Pressure Management
and Control (PRESMAC) model which has been developed through the Water Research
Commission (WRC) funded project titled “The Water Leakage: Pressure Management
Model”.
The PRESMAC model represents one of several models that are being developed
through the WRC in order to assist water suppliers to manage and reduce their levels of
unaccounted-for water. The models are supplied free-of-charge through the WRC for use
within South Africa and further details can be obtained from the WRC web site on:
http://www.wrc.org.za.
DISCLAIMER
Every effort has been taken to ensure that the model and manual are accurate and
reliable. Neither the Water Research Commission nor the model developers
(R McKenzie, A Lambert), shall, however, assume any liability of any kind resulting from
the use of the program. Any person making use of the PRESMAC model, does so entirely
at his/her own risk.
COPYRIGHT
The model and manual have been developed through the South African Water Research
Commission (WRC). The WRC encourages the use and dissemination of information and
software emanating from their research projects and the duplication and re-distribution of
this software is therefore permitted. Similarly, duplication and re-distribution of the user-
manual is also permitted provided that due recognition is given to both the WRC and the
developers. All copies of the software and manual should be attended by the above
disclaimer.
TECHNICAL SUPPORT
The WRC does not provide technical support on the PRESMAC model and any questions
or problems associated with the program can be directed to the model developers at
ronniem@wrp.co.za or wrp@wrp.co.za.
iii 02/05/16
Pressure Management Program (PRESMAC)
Executive Summary
The Problem
In the continual battle to reduce leakage from potable water distribution systems, the
influence of pressure is often overlooked. Planners design potable water distribution
systems to provide a certain minimum level of service (usually in the order of 25 m of
pressure) throughout the day at the most critical point in the system. The critical point is
generally either the highest point in the system or the point most distant from the source,
although it may be a combination of the two depending upon local topography.
The pressure at the critical point will depend upon the pressure at the inlet point minus the
friction losses occurring between the inlet and the critical points. The friction losses will be
highest during periods of peak demand; typically during the breakfast period and again
during the early evening period when most consumers are using water for washing,
cooking, gardening etc. After the evening peak, the pressure throughout the system will
gradually increase due to reduced friction losses and, in certain cases, also the filling up
of local storage reservoirs.
Since the systems are designed to supply the minimum level of pressure at the critical
point during the peak demand periods, it is clear that the pressure will increase during the
periods of low demand. The pressures in potable water distribution systems are therefore
significantly higher than required much of the time, particularly during the night when most
of the consumers are sleeping. Since losses and leakage from a system are highly
dependent upon pressure, it is also clear that leakage rates will be highest during the
periods when few, if any, consumers wish to use water.
Software Solutions
Although there is no simple solution to the complex problem of excess pressure in a water
distribution system, considerable research and development has taken place over the
past decade. This has resulted in the creation of various techniques and equipment that
can help to control pressure and, thus, reduce leakage.
iv 02/05/16
In 1991, a National Leakage Initiative was established in the UK by the Water Services
Association and the Water Companies Association to update and review the guidelines
concerning leakage control that had been in use since 1980. It was agreed by all
organisations involved in potable water supply that the guidelines required updating in
view of the considerable progress that had been made over the previous ten-year period.
As a result of new water legislation, it became necessary for all water suppliers to
demonstrate to the regulators that they fully understood their position on leakage. This
did not imply that all water suppliers had to demonstrate the lowest achievable leakage
levels, but simply that they were applying correct and appropriate economic and resource
principles. To this end, it was agreed that all water suppliers would adopt a
straightforward and pragmatic approach to leakage levels. This was achieved through the
development of various techniques that became known as the Burst and Background
Estimate (BABE) procedures.
The BABE procedures were developed over a period of approximately 4 years by a group
of specialists selected from several of the major water supply companies based in
England and Wales. The group was instructed to develop a systematic and pragmatic
approach to leakage management that could be applied equally well to all of the UK water
supply utilities. The result of this initiative was a set of 9 reports published by the UK
Water Industry (WRC) on the subject of managing leakage. The nine WRC reports cover
the following topics :
The intention of the reports was not to be prescriptive, but to provide a “tool kit” to the
water industry to enable the water supply managers to evaluate leakage levels and to
manage the system.
Pressure management was identified as one of the key issues with the result that one full
report was dedicated to the subject (Report G). The main problem was to develop a
simple and pragmatic approach to predicting the reduction in leakage that can be
achieved through a range of possible pressure management measures. Several of the
UK water companies developed commercial software to address this problem.
Hardware Solutions
At the same time as the research into pressure management was being completed,
several new pressure controllers were also being developed which were able to modulate
v 02/05/16
the pressure at a pressure reducing valve (PRV) situated at the inlet to a pressure zone.
By using such controllers it became possible to reduce the pressure during periods of low
demand and reduce leakage without adversely affecting the level of service to the
consumers. For the first time ever, both software and hardware solutions could be used
to tackle pressure in potable water distribution systems.
In general, there are two types of Advanced Pressure Control: time-modulated control and
flow-modulated control. The time-modulated controller offers the simplest and the least
expensive form of Advanced Pressure Control. It is basically a timing device that can be
attached to the controlling pilot on any normal PRV to reduce the outlet pressure at certain
times of the day. It is a very simple and compact device that can accommodate up to four
switching periods each day and two pressure levels: a high-pressure setting dictated by
the PRV and a low-pressure setting as adjusted on the controller. The time-modulated
controller is simple and easy to use and represents the least expensive form of advanced
pressure control. It has certain limitations; one of which concerns the influence on fire-
fighting flows. If fire-fighting flows present a problem, the time-modulated option may not
be suitable; in which case the more advanced flow-modulated controller may be required.
The second and more complex controller is the flow-modulated controller which provides
greater flexibility and control than that offered by the simpler time-modulated controller.
The flow-modulated controller will control the pressure at the inlet point in accordance with
the demand being placed on the system. During peak demand periods, the maximum
pressure as dictated by the PRV will be provided, while at low demand periods the
pressure will be reduced to minimise excess pressure and the associated leakage. The
flow-modulated controller can be equipped with a telephone or radio link to the critical
point and, in this manner, the inlet pressure can be adjusted to ensure that there is
virtually no excess pressure at the critical point at any time. This will then provide the
minimum leakage achievable. Although the flow-modulated controller is more expensive
than the simpler time-modulated controller, it does offer greater flexibility which can be
important in certain areas where fire-fighting requirements represent a potential problem.
The PRESMAC Model
Although the pressure management software developed in the UK is available

commercially to companies or consultants throughout the world, it is not designed
specifically for South African conditions. In addition, this software is relatively expensive
in rand terms overseas. Although the potential savings can be very significant, many of
vi 02/05/16
the smaller municipalities are unable to budget for such software without demonstrating
the savings in advance – clearly a cart and horse situation.
To overcome these problems and as part of a greater strategy by the South African Water
Research Commission to promote water conservation, a project was initiated in 1999 to
develop a South African pressure management model (PRESMAC). The new model is
based on the same BABE principles but it was modified to suit South African conditions
where necessary. As opposed to the UK models which are based on the EXCEL
spreadsheet architecture, the new South African model is written in DELPHI and was
developed locally with support from Mr Allan Lambert.
The PRESMAC pressure management model is used to assess the likely savings (in
monetary terms) of various pressure reduction options (fixed-outlet and time-modulated
PRV’s) in a selected zone metered area. The analysis is undertaken in a relatively simple
and pragmatic manner based on the general BABE concepts. This approach allows the
user of the program to gauge the potential for pressure management very quickly and
effectively without requiring a full detailed pipe network analysis. Although the
methodology is based on a number of simplifications and assumptions, in practice, the
predicted savings are generally within 10% to 20% of those actually achieved (erring on
the conservative side).
Data Requirements
To use the PRESMAC model the user must collect certain basic information for the zone
metered area or pressure management area in question. The basic information required
includes:
• number of connections;
• length of mains;
• number of properties;
• population;
• expected leakage rates from connections, properties and mains;
• pressure exponent for the system as a whole;
• details of any commercial consumers.
vii 02/05/16
The information used in PRESMAC is basically the same information used in a normal
minimum nightflow analysis. In addition to the basic information, however, the user must
provide three 24-hour pressure profiles and also the 24-hour zone inflow. The average
hourly values are required at the following points:
• pressure at the inlet point;
• pressure at the average zone point;
• pressure at the critical point;
• inflow to the zone.
These four sets of hourly values are usually measured using flow and pressure loggers
which are attached to the flow meter at the zone inlet as well as at three other suitable
points. The pressures can basically be logged at any suitable point such as a fire hydrant
or a tap located in/on someone’s property.
Using PRESMAC
The PRESMAC model allows the user to analyse the existing situation in any specific
pressure management area. It then allows the user to assess the likely savings that can
be achieved through the installation of a new PRV or by re-setting an existing PRV to a
lower pressure. Finally, the model allows the user to assess the potential savings that can
be achieved through the use of a time-modulated controller. The time-modulated
controller is the simplest, least expensive and most widely used controller available. It is
already in use in many parts of South Africa having been introduced to the country at the
beginning of 1999.
It should be noted that the model in its current form does not accommodate the analysis of
the more complicated and expensive flow-modulated controller, although this option may
be added at some future date. In most cases, the analysis of the time-modulated
controller will provide the required motivation for the purchase and installation of any form
of advanced pressure control. If it is found that the time-modulated controller can be
justified on sound financial grounds, then it is likely that the flow-modulated controller will
provide even greater savings.
viii 02/05/16
PRESMAC
Pressure Management Program
Table of Contents
Page No.
EXECUTIVE SUMMARY iv
1. PRESSURE MANAGEMENT MODEL ............................................................................1—1
1.1. GENERAL CONCEPTS OF PRESSURE MANAGE MENT....................................................1—1
1.2. CONCEPTS OF ACTIVE PRESSURE CONTROL................................................................1—3
1.3. THE PRESMAC MODEL.......................................................................................................1—10

1.3.1. General 1—10
1.3.2. Data Requirements................................................................................................................1—11
1.3.3. Analysis Approach.................................................................................................................1—12
1.4. METHODOLOGY ..................................................................................................................1—17
1.5. PURPOSE AND LAYOUT OF THIS REPORT.....................................................................1—18
2. DETAILS OF THE PRESMAC MODEL...........................................................................2—1
2.1. HOW THE MODEL W ORKS...................................................................................................2—1
2.2. PRESSURE EXPONENT (N1)................................................................................................2—2
2.3. CALCULATION OF THE HOUR-DAY FACTOR ....................................................................2—4
2.4. PRESSURE-DEPENDENT AND PRESSURE-INDEPENDENT FLOWS..............................2—5
2.5. ESTIMATION OF NETWORK FRICTION (“K”) FACTORS....................................................2—9
2.6. ANALYSIS OF FIXED-OUTLET PRV...................................................................................2—11
2.7. ANALYSIS OF TIME-MODULATED CONTROLLER...........................................................2—18
3. USING PRESMAC ..............................................................................................................3—1
3.1. HARDWARE AND SOFTWARE REQUIREMENTS...............................................................3—1
3.2. INSTALLING PRESMAC.........................................................................................................3—1
ix 02/05/16
3.3. DATA REQUIREMENTS.........................................................................................................3—1
3.4. RUNNING PRESMAC .............................................................................................................3—2
3.5. DATA INPUT ...........................................................................................................................3—3

3.5.1. General_Info ............................................................................................................................3—4
3.5.2. Default Parameters..................................................................................................................3—4
3.5.3. Flow and Pressure...................................................................................................................3—6
3.5.4. N1 Calculations .......................................................................................................................3—7
3.6. CURRENT SITUATION...........................................................................................................3—9
3.7. FIXED-OUTLET PRV............................................................................................................3—12
3.8. TIME-MODULATED PRV......................................................................................................3—13
3.9. SUMMARY ............................................................................................................................3—14
4. REFERENCES.....................................................................................................................4—1
FIGURES
Figure 1.1 Typical zone pressure distribution during peak demand periods
Figure 1.2 Typical zone pressure distribution during low demand periods
Figure 1.3 Pressure control using conventional PRV
Figure 1.4 Typical 4-point time modulated pressure profile
Figure 1.5 Typical installation of a time-modulated PRV controller
Figure 1.6 Pressure control using a time-modulated PRV controller
Figure 1.7 Components required for a flow modulated PRV installation
Figure 1.8 Pressure control using a flow-modulated PRV controller
Fibure 1.9 Pressure control using a telemetry linked flow-modulated PRV
controller
Figure 1.10 Typical zone with no pressure control
Figure 1.11 Typical zone with fixed-outlet pressure control
Figure 1.12 Typical zone with time-modulated pressure control
Figure 1.13 Typical zone with flow-modulated pressure control
Figure 2.1 Example of a (two-point) time-modulated pressure profile)
Figure 3.1 Opening banner from PRESMAC
Figure 3.2 Main menu in PRESMAC
Figure 3.3 The Default_Parameters Form (from the Data_Input Form)
Figure 3.4 Details of the Flow and Pressure Form
Figure 3.5 Example of a Pressure Step-Test Analysis
x 02/05/16
Figure 3.6 Tabulated Information from the Current_Situation Form

Figure 3.7 Graphical Output from the Current_Situation Form
Figure 3.8 Graph options available
Figure 3.9 Typical data entry form for the Fixed Outlet PRV Option
Figure 3.10 Example of the Time-Modulated PRV Form
Figure 3.11 Example of the Summary Form
Figure 3.12 Example of “Implementation Costs” Form
TABLES
Table 2.1 Calculation of N1 from pressure step-test analysis
Table 2.2 Calculation of the N1 values for example shown in Table 2.1
Table 2.3 Estimation of pressure-independent flow at MNF (item 1 in
Table 2.6)
Table 2.4 Estimation of pressure-dependent flow at MNF (item 2 in
Table 2.6)
Table 2.5 Estimation of pressure-dependent flow at hour 0 – 1 (item 3 in
Table 2.6)
Table 2.6 Example to demonstrate the calculation of pressure-dependent and
pressure-independent flow
Table 2.7 K factors for AZP and critical points (items 2 and 3 in Table 2.8)
Table 2.8 Estimation of K-values at AZP and critical point
Table 2.9 Iterative approach used to calculate the new pressure at the AZP
Table 2.10 Revised pressures at AZP and critical points for fixed outlet PRV
Table 2.11 Derivation of new inlet pressure to provide critical pressure of 10 m
Table 2.12 Savings achieved through the use of PRV set at 38.5 m
APPENDIX A: Introduction to BABE and FAVAD Concepts, and Calculation

of Unavoidable Annual Real Losses
APPENDIX B: Data Capture Form for PRESMAC
xi 02/05/16
1. PRESSURE MANAGEMENT MODEL
1.1. GENERAL CONCEPTS OF PRESSURE MANAGEMENT
Most water reticulation systems are designed to provide a minimum working pressure at
all points in the system throughout the day. This means that the minimum pressure
(normally specified in the local by-laws) occurs at some critical point in the system which
is often either the highest point in the system or the point furthest from the supply.
Most water distribution systems experience significant fluctuations in demand throughout

the day with morning and evening peaks coupled with periods of low demand during the
night and sometimes also during the early afternoons. Many systems also experience
seasonal fluctuations caused by climatic factors that influence irrigation requirements or
by holiday migration that can significantly influence the demand for periods of days or
weeks at a time.
Since most systems are designed to provide a set minimum pressure throughout the day,
they are generally designed to meet this pressure requirement during periods of peak
demand when the friction losses are at their highest and inlet pressures are at their
lowest. As a result of this design methodology, most systems experience higher
pressures than necessary during the remaining non-peak demand periods. This is
evident from the fact that in most areas the major burst pipes tend to occur during the late
evening and early morning periods when system pressures are at their highest.
This concept is shown graphically in Fig. 1.1 which represents a typical pressure situation
for a zone at peak demand periods where the minimum pressure required is 20 m.
The same zone is shown again in Fig. 1.2 for periods of low demand, typically
experienced during the late evening and early hours of the morning (assuming that the
properties use direct feeds with little or no roof storage).
1—1 02/05/16
Just
upstream
of PRV
100
Just
downstream
of PRV
At critical
point
60
20
PRV
20 = pressure in m
WRP_006.cdr
Figure 1.1: Typical zone pressure distribution during peak demand periods
Just
upstream
of PRV
120
Just
downstream At critical
of PRV point
60 60
PRV
20 = pressure in m
WRP_007.cdr
Figure 1.2: Typical zone pressure distribution during low demand periods
From Figs. 1.1 and 1.2 it can be appreciated that for most of the time the pressure in a
water distribution system is likely to be considerably higher than required (unless some
form of active pressure management has already been implemented). If it is also accepted
that leakage increases with increased pressure, then it can be concluded that leakage
levels in most systems are higher than they should be during most of the time.
1—2 02/05/16
It is clear that if the excess pressure in a system can be reduced, then so too can the
leakage, which, in turn, will save money. This is the basic philosophy governing pressure
management in potable water distribution systems and is often referred to as “Active
Pressure Control” or “Advanced Pressure Control”.
1.2. CONCEPTS OF ACTIVE PRESSURE CONTROL

The main objective of active pressure control is to minimise the excess pressure in a
water distribution system which, in turn, will reduce leakage as well as the frequency of
burst pipes. This simple objective is often difficult to achieve in practice due to numerous
external factors that must be taken into account such as fire-fighting requirements, high-
rise buildings etc. In general, however, significant savings can often be made and there
are many examples throughout the world (and now also in South Africa) where active
pressure control has been extremely successful.
It should be noted that there is often a misconception that pressure control is aimed at
reducing the levels of service to the consumer. While pressure management can be used
to reduce customer demand, this is generally not the primary objective. As mentioned
above, the main objective is to reduce the “excess pressure” during periods of low
demand. If this can be achieved through proper and careful pressure management
measures, it should be possible to reduce leakage and burst frequency without any
detrimental effect to either the consumer or the fire-fighting services. Obviously there are
numerous potential problems and pit-falls. However, through experienced planning it
should be possible to overcome most of these.
Although there is no simple solution to the complex problem of excess pressure in a water
distribution system, considerable research and development has taken place over the
past decade. This has resulted in the creation of various techniques and equipment that
can help to control pressure and to reduce leakage.
In 1991 a National Leakage Initiative was established in the UK by the Water Services
Association and the Water Companies Association to update and review the guidelines
concerning leakage control that had been in use since 1980. It was agreed by all
organisations involved in potable water supply that the guidelines required updating in
view of the considerable progress that had been made over the previous ten-year period.
To this end, it was agreed that all water suppliers would adopt a straightforward and
pragmatic approach to leakage management. This was achieved through the
1—3 02/05/16
development of various techniques that became known as the Burst and Background
Estimate (BABE) procedures.
The BABE procedures were developed over a period of approximately 4 years by a group
of specialists selected from several of the major water supply companies based in
England and Wales. The group was instructed to develop a systematic and pragmatic
approach to leakage management that could be applied equally well to all of the UK water
supply utilities. The result of this initiative was a set of 9 reports published by the UK
Water Industry (1994) on the subject of managing leakage.
The intention of the reports was not to be prescriptive, but to provide a “tool kit” to the
water industry to enable the water supply managers to evaluate leakage levels and to
manage the system. Pressure management was identified as one of the key issues with
the result that one full report was dedicated to the subject (Report G). The main issue
addressed in the report was the development of a simple but realistic approach to
predicting the reduction in leakage that can be achieved through a range of possible
pressure management measures. Several of the UK water companies subsequently
developed commercial software to address this problem based on the methodology
outlined in the report. This software has since been used in many parts of the world
including several parts of Europe, Brazil, Ghana, South Africa and Malaysia.
At the same time as the research into pressure management was being completed,
several new pressure control devices were also being developed which were able to
modulate the pressure at a pressure reducing valve (PRV), based on either time of day or
the flow through the valve. By using such controllers it became possible to reduce the
pressure during periods of low demand and thus reduce leakage without adversely
affecting the level of service to the consumers. It should be noted that there are various
other techniques of achieving the same goals and several of the large valve
manufacturers have developed their own techniques, many of which are hydraulically
based. For the purpose of this project, however, only the electronic controllers are
considered since they dominate the PRV control market and have been used successfully
in many parts of South Africa.
With the aid of the new software and the use of the new PRV controllers, it became
possible, for the first time, to accurately assess the potential savings that can be achieved
from the various pressure management options. In this manner the savings can first be
1—4 02/05/16
estimated and then used to motivate the implementation of the physical devices. It also
prevents the installation of expensive equipment in cases where it will not be cost-
effective.
There are several types of PRV controllers available both electrically-operated and
hydraulically-operated. For the purpose of the study, three possible forms of pressure
control were considered of which the first two are incorporated into the PRESMAC Model.
• Fixed outlet PRV;
• Time-modulated PRV;
• Flow-modulated PRV.
The first option is simply a normal PRV which is used to provide a continuous pressure at
the inlet to a zone as shown in Fig. 1.3.
Day 1 Day 2 Day 3 Day 1 Day 2 Day 3
time time
Inlet Outlet
Extremity
Pressure reducing
valve District load
WRP_005.cdr
Figure 1.3: Pressure control using conventional PRV
Time-modulated controller
The time-modulated controller is the simplest form of Advanced Pressure Control and also
the least expensive. It is a timing device that can be attached to the controlling pilot on
1—5 02/05/16
any normal PRV to reduce the outlet pressure at certain times of the day. It is a very
simple and compact device that can accommodate four switching periods each day and
two pressure levels: a high level dictated by the PRV itself and a low level as set on the
controller. This is a simple but effective method of reducing pressures in systems where
there is some consistent pattern of demand on a daily basis. It is an ideal solution for
reducing excessive pressures at night when most of the consumers are asleep and the
demand for water is minimal. In such cases the night-time pressure can often be reduced
significantly without lowering the normal levels of service to the consumers.
Up to two time periods can be specified (see Fig. 1.4) per day although, in most cases,
only one is needed. A typical installation of a time-modulated controller is shown in
Fig. 1.5 and the general objectives are shown in Fig. 1.6.
Set on PRV
High P2 P4
Pressure
Set on
Controller
Low P1 P3
WRP_002.cdr
0 2 4 6 8 10 12 14 16 18 20 22 24
Time (hours)
Figure 1.4: Typical 4-point time modulated pressure profile
1—6 02/05/16
Photograph courtesy
Pressure Management Systems
And Krugersdorp TLC
And Krugersdorp TLC
Figure 1.5: Typical installation of a time -modulated PRV controller
Day 1 Day 2 Day 3 Day 1 Day 2 Day 3
Time Modulated time time

Controller
Inlet Outlet Extremity
Pressure reducing
valve
District load
WRP_012.cdr
Figure 1.6: Pressure control using a time -modulated PRV controller
1—7 02/05/16
It should be noted that the time-modulated controller shown in Fig. 1.5 (see arrow) is a
simple and self-powered unit which can operate for approximately 5 years on a single
battery. It is programmed through the use of two buttons on the fascia, much in the same
way as one sets a normal watch or alarm clock.
The main application of the time-modulated PRV is to reduce pressures during periods of
low demand when the system pressures tend to be higher than necessary, resulting in
excessive pressures at the critical point. Through the use of such a controller, it is
possible to cut out some of the high-pressure peaks especially during night-time periods.
The main potential problem with the time-modulated controller concerns the fire-fighting
requirements. The controller cannot react to an increase in demand caused by the
opening of a fire hydrant with the result that there can be problems if a fire breaks out
during the period of low pressure. In many parts of South Africa, however, it appears from
discussions with various fire departments that this is not a problem since there are either
no fire hydrants or they have been vandalised to the extent that they are inoperable.
Under such conditions, the fire departments bring in their own water and do not try to use
the fire hydrants even if they are available. Another limitation of the time-modulated
controller is that the pressure difference between the high and low settings should ideally
not exceed 20 m, otherwise water hammer and/or cavitation may become problem issues.
Flow-Modulated Controller
The second and more complex controller is the flow-modulated controller which provides
greater flexibility and control than that offered by the simpler time-modulated controller.
Unfortunately, the greater flexibility is accompanied by a higher cost and the flow-
modulated controller is approximately double the cost of the time-modulated version. The
typical components required for a flow-modulated PRV installation are shown in Fig. 1.7.
1—8 02/05/16
Photograph courtesy of Technolog, UK.
Figure 1.7 : Components required for a flow-modulated PRV installation
The flow-modulated controller will control the pressure at the inlet point in accordance with
the demand being placed on the system. During peak demand periods, the maximum
pressure as dictated by the PRV will be provided, while at low demand periods the
pressure will be reduced to minimise excess pressure and the associated leakage. The
concepts of the flow-modulated controller are shown in Fig. 1.8.
Flow Modulated Day 1 Day 2 Day 3 Day 1 Day 2 Day 3
Controller time time
PRVController
Inlet Outlet Extremity
Water Pressure
District load
meter reducing valve WRP_011.cdr
Figure 1.8: Pressure control using a flow-modulated PRV controller
1—9 02/05/16
The flow-modulated controller can easily be equipped with a telephone or radio link to the
critical point and, in this manner, the inlet pressure can be adjusted to ensure that there is
virtually no excess pressure at the critical point at any time throughout the day. This will
provide the most effective control possible (without reducing the size of the zone) and is
depicted in Fig. 1.9
Radio or
Controller
leased line link
Pressure sensor
and modem
Inlet
Pressure Critical node

Water supply
reducing valve district
WRP_003.cdr
Figure 1.9: Pressure control using a telemetry linked flow-modulated PRV controller
1.3. THE PRESMAC MODEL
1.3.1. General
Although the pressure management software developed in the UK is commercially
available to companies or consultants throughout the world, it is not designed specifically
for South African conditions, nor is it supported by any organisation in South Africa.
Although the potential savings can be very significant, many of the smaller water suppliers
in South Africa are unable to budget for such software without demonstrating the savings
in advance – clearly a cart and horse situation.
To overcome these problems and as part of a greater strategy by the South African Water
Research Commission to promote water conservation, a project was initiated in 1999 to
develop a South African pressure management model (PRESMAC). The model is based
on the same BABE principles as the existing UK models and modified to suit South
1—10 02/05/16
African conditions where necessary. As opposed to the UK models which were based on
the EXCEL spreadsheet architecture, the new South African model is written in DELPHI
and was developed locally with support from Bristol Water Consultancy Services and Mr
Allan Lambert who were both instrumental in the development and use of the UK models
(Lambert et al, 1998).
The PRESMAC pressure management model is used to assess the likely savings (in
monetary terms) of various pressure reduction options in a selected zone metered area.
This approach allows the user of the program to gauge the potential for pressure
management very quickly and effectively without undertaking a full detailed pipe network
analysis. Although the methodology is based on a number of simplifications and
assumptions, in practice the predicted savings are generally within 10% to 20% of those
actually achieved (erring on the conservative side).
1.3.2. Data Requirements
To use the PRESMAC model the user must collect certain basic information for the zone
metered area or pressure management area in question. The basic information required
includes:
• number of connections;
• length of mains;
• number of properties;
• population;
• expected leakage rates from connections, properties and mains;
• pressure exponent for the system as a whole;
• details of any commercial consumers.
The information used in PRESMAC is basically the same information used in a normal
minimum nightflow analysis and all of the items mentioned above are explained fully in the
SANFLOW user guide (WRC, 1999). In addition to the basic information, however, the
user must provide three 24-hour pressure profiles and also the 24-hour zone inflow. The
average hourly values are therefore required at the following points:
1—11 02/05/16
• pressure at the inlet point,
• pressure at the average zone point,
• pressure at the critical point,
• inflow to the zone.
These four sets of hourly values are usually measured using flow and pressure loggers
which are attached to the flow meter at the zone inlet as well as at three other suitable
points such as a fire hydrant or a tap located on someone’s property. The information
required to run the PRESMAC Model is shown in the data input form in Appendix B.
1.3.3. Analysis Approach
The analysis approach involves investigating each area carefully before any new
equipment is installed. Each zone should be checked for integrity since any results
obtained from a non-discrete zone will be questionable. It should be noted that, although
preferable, it is not always necessary to have a single supply point to a particular zone. In
cases where there are two or even three supply points, it is still possible to carry out a
meaningful analysis as long as all of the supply points are monitored simultaneously.
After logging the zone inflows and pressures at the various key points, the pressure
analysis program (PRESMAC) can be used to assess the scope (if any) for reducing
leakage through pressure control. From the results, the appropriate equipment can then
be selected and installed after which follow-up loggings should be undertaken to verify the
results. It should be noted that it is usually necessary to allow for some additional time to
“calibrate” the controller after it has been installed. Although the PRESMAC Model will
provide an indication of the pressure control limits based on the 24-hour pressure profiles
supplied from the logging exercise, these limits often have to be adjusted in some way
since most systems do not follow the exact profile provided each day of the week. As a
result, there is always some fine tuning and manual adjustment required to obtain the best
results without any consumer complaints.
Fig. 1.10 shows a typical situation encountered in a relatively small and well-managed
supply zone. The figure shows the inlet pressure, the zone inflow and the pressure at the
critical point during a 24-hour period.
1—12 02/05/16
40 90
Inlet Pressure Pressure at 80

35 Critical Point
70
30
60
Pressure (m)
Flow (m /hr)
25
3
50
20
40
15
30
Minimum
Acceptable
10 Pressure 20
Zone Inflow
5 Excess pressure 10
at critical point
WRP13_Fig14
0 0
0 2 4 6 8 10 12 14 16 18 20 22 24
Time (hours)
Figure 1.10: Typical zone with no pressure control
As can be seen, the inlet pressure is not particularly high, although there is some scope
for improvement. It can also be seen from the figure that the zone inflow exhibits the
typical daily peak demand pattern found throughout South Africa with high demand
periods in the morning and early evening. Virtually all of the consumers are supplied
directly from the water mains and there are few, if any, storage tanks in the area.
For the purpose of this example, it has been decided that the minimum acceptable
pressure throughout the day is 15 m. This minimum pressure has been selected for
illustrative purposes and does not indicate that 15 m should be accepted as the minimum
pressure in all zones. In some zones, the fire-fighting requirements may be 30 m, in
which case there would be virtually no scope for further pressure management. In the
example, however, it can be seen from the pressure at the critical point that there is
considerable excess pressure in the system as indicated by the shaded portion in the
figure.
From Fig. 1.10 it can also be seen that neither the pressures nor the minimum night flow
are unusually high. There is obviously some room for improvement. However, the
example is typical of a normal zone that may be found in many parts of the world. The
savings in such a zone are unlikely to be as spectacular as other examples often quoted
from various parts of Africa, Brazil and Malaysia etc. As a general rule of thumb, it is
1—13 02/05/16
usually possible to reduce the minimum night flow by around 50% in cases where it is
already high due mainly to background leakage and/or internal plumbing leakage. If the
minimum night flow is less than 20 m3/hr, it is unlikely that the savings achieved will be
sufficient to pay for the installation in less than one or two years. Ideally, pressure
management should be considered as an option in zones where the minimum night flow is
above 20 m 3/h and preferably above 50 m 3/h.
As can be seen from the zone inlet pressure in Fig. 1.10, the zone considered in the
example has no form of pressure control and, therefore, the first option to be analysed is a
standard fixed outlet PRV.
The PRESMAC Model is used to analyse the situation when a standard PRV is installed
and set to an inlet pressure that will just provide 15 m of pressure at the critical point
during the peak demand period. The corresponding figure is shown in Fig. 1.11 from
which it can be seen that a fixed outlet PRV can be used to ensure that the pressure at
the critical point is limited to the minimum required pressure (15 m in this example) at the
period of maximum demand. It can also be seen from the figure, however, that there
remains considerable excess pressure in the system during most of the day.
To reduce the excess pressure it is necessary to use a more sophisticated form of

pressure control than simply a standard PRV. As mentioned previously, there are many
types of PRV controllers available and most large valve manufacturers have several
products which can be used to reduce excess pressure in a system. One such product is
a simple yet effective time-modulated controller which has been used with great success
in many parts of the world including South Africa. Typical results from the model for the
time-modulated option are shown in Fig. 1.12.
From Fig. 1.12 it can be seen that the use of the time-modulated controller results in a
significant reduction in the excess system pressure as indicated by the smaller shaded
portion in the figure.
If further improvement is still required, the more sophisticated flow-modulated controller

can be considered. The results from the flow-modulated controller are shown in Fig. 1.13
from which it can be seen that there is some improvement on the results depicted in
Fig. 1.12. It should be noted that the PRESMAC Model does not offer the option of
analysing the flow-modulated controller. Details of the flow-modulated controller have
been included for completeness and to demonstrate that further improvements can often
be achieved over those produced by a time-modulated controller. For analysis purposes,
1—14 02/05/16
it is usually sufficient to model the time-modulated controller to determine if there is scope

for pressure control after which the controller can be selected based on the local
conditions and requirements.
40 90
35 80
Pressure at
Inlet Pressure Critical Point 70
30
60
Pressure (m)
Flow (m /hr)
25
3
50
20
40
15
30
Minimum
Acceptable
10 Pressure
20
Zone Inflow
5 Excess pressure 10
at critical point
WRP13_Fig14
0 0
0 2 4 6 8 10 12 14 16 18 20 22 24
Time (hours)
Figure 1.11: Typical zone with fixed-outlet pressure control
40 90
35 80
Pressure at
Inlet Pressure Critical Point
70
30
60
Pressure (m)
Flow (m /hr)
25
3
50
20
40
15
30
Minimum
10 Acceptable
Pressure 20
5 Zone Inflow Excess pressure 10

at critical point
WRP13_Fig14
0 0
0 2 4 6 8 10 12 14 16 18 20 22 24
Time (hours)
Figure 1.12: Typical zone with time -modulated pressure control
1—15 02/05/16
40 90
35 80
Pressure at
Inlet Pressure Critical Point 70
30
60
Pressure (m)
Flow (m /hr)
25
3
50
20
40
15
30
Minimum
Acceptable
10 Pressure 20
5 Zone Inflow Excess pressure 10

at critical point
WRP13_Fig14
0 0
0 2 4 6 8 10 12 14 16 18 20 22 24
Time (hours)
Figure 1.13: Typical zone with flow-modulated pressure control
In this particular example, there is little additional financial benefit to be gained from the
use of the flow-modulated controller over the cheaper and less sophisticated flow-
modulated controller. If the fire-fighting requirement is a potential problem, then selection
of the controller may be dictated by functionality rather than economy. In some cases
there is little benefit to be gained from even the time-modulated controller and a standard
PRV without advanced pressure control may then be the most appropriate measure.
The decision regarding which form of PRV controller to use (if any) in a particular system
is generally based on sound financial considerations with the proviso that the equipment
selected also satisfies local fire-fighting regulations. By using the computer model to
predict the savings through reduced leakage associated with the various pressure
management options, it is possible to estimate the payback periods for each option. This
information can then be used to select the most appropriate option. It should be noted
that the option with the shortest pay-back period is not always selected since there are
often external considerations which dictate the final choice as previously mentioned.
1—16 02/05/16
1.4. METHODOLOGY
To carry out the analysis using the Pressure Management Model (PRESMAC) two basic
equations are used.
Equation 1: Pressure/Loss Equation
L0/L1 = (P0/P1)N Eq. 1(a)
or
L1 = L0 * (P1/P0)N Eq. 1(b)
where
L0 = initial leakage loss in m 3/h
L1 = new leakage loss in m 3/h
P0 = initial pressure (m)
P1 = new pressure (m)
N = pressure exponent (non-dimensional) – see Section 2.2
The above equation can also be solved for N to give
N = log (L0 / L1) / log (P0 / P1) Eq. 1(c)
Equation 2: Head Loss/Flow equation
The second equation used in the PRESMAC Model is basically a head loss equation used
to estimate the head loss between the inlet point and both the AZP and critical points for
any particular flow. It is a simplification of the normal friction factor equations in which all
of the terms excluding the flow are lumped into a single coefficient K. The equation used
is shown in Equation 2.
HL = K * Q2 Eq. 2
1—17 02/05/16
where
HL = head loss in m
K = head loss coefficient (m-5.h2)
Q = flow in (m3/h)
Using the two equations and the BABE methodology, the PRESMAC model can be used
to analyse and test the various pressure management options.
1.5. PURPOSE AND LAYOUT OF THIS REPORT
The purpose of this report is to explain the methodology behind the new PRESMAC
Model and to provide a simple and clear user manual to the model. The report is set out
in 5 sections, details of which are provided below.
Section 1: Introduction and General

This section provides a general introduction to the concepts of Active Pressure Control
and explains the need for and use of the new PRESMAC Model. Some details explaining
the basic concepts upon which the PRESMAC Model is based are also provided, although
no in-depth theory is given.
Section 2: Details of the PRESMAC Model

This section provides details of all variables used in the PRESMAC Model as well as the
various items of information that are generated by the model. It effectively explains how
the model functions and provides more detailed information on certain key issues that
must be clearly understood when using the model. This section should be consulted by
all users.
Section 3: Using the PRESMAC Model

This section is effectively the “User Guide” to the PRESMAC Model and is possibly one of
the most important sections in this report. Hardware and software requirements are
discussed as well as the installation and running procedures for the model. The user is
guided step-by-step through the installation and startup procedures while all windows and
associated forms are explained in detail.
Section 4 : References
This section provides a few useful references for users wishing to gain more in-depth
knowledge of the subject of pressure management. Sufficient information is already
1—18 02/05/16
provided in the report for most users and the references will only be needed by those who
wish to gain a more comprehensive understanding of advanced pressure control or the
BABE procedures.
Appendix A : Introduction to BABE and FIVAD Concepts

This Appendix provides details of the basic concepts of the BABE procedures which are a
key element to all of the water demand management programs being developed through
the Water Research Commission (WRC). This section is repeated in the reports for each
of the four models developed for the WRC and need not be consulted if the user has
already studied the concepts in a previous report. In order to use the PRESMAC Model
properly it is important that the user has a firm understanding of the BABE procedures and
all new users should familiarise themselves with Appendix A before proceeding. This
appendix provides details of the basic Burst and Background Estimate concepts as well
as details of the Fixed and Variable Area Discharge concepts upon which the PRESMAC
Model is based.
Appendix B: Data entry forms and Sample Data Set

Appendix B contains an example of a data entry form which can be used to capture the
basic information required to run the PRESMAC Model. A blank form as well as a typical
completed form are provided for reference. A sample data file is also included in
Appendix B for reference.
1—19 02/05/16
2. DETAILS OF THE PRESMAC MODEL
2.1. HOW THE MODEL WORKS

Before explaining the operation of the PRESMAC Model it is important to appreciate that
no two water supply systems are identical and that the water lost through leakage in a
particular system is dependent upon many factors including:
• The length of mains
• The number of service connections;
• The average operating pressure;
• The pipe material and ground conditions;
• The quality of workmanship when the system was installed;
• The levels of cathodic protection in the case of iron and steel pipes;
• Various other factors.
Obviously it is not possible to take everything into account and in most cases it is difficult
to obtain even the most basic information concerning the pipe network. It is also known
that the leakage from any system is a combination of burst leakage and background
leakage, both of which react differently to changes in pressure. For the purpose of the
PRESMAC Model, various simplifications are adopted which are considered to be realistic
in view of the many unknown factors that influence the leakage predictions. The most
important assumptions and issues are discussed separately in the remainder of Section 2
of this report.
In order to assess the leakage occurring in a particular water distribution system, the
PRESMAC Model undertakes a series of calculations in a step-wise manner:
Step 1: The measured zone inflow is first split into pressure-dependent

and pressure-independent components that will naturally react
differently to changes in pressure.
Step 2: Calculate the K factors (as mentioned in Eq. 2) for the head
losses between the inlet point and the Average Zone Point
(AZP) as well as between the inlet point and the critical point for
each hour in the 24-hour analysis period.
Step 3: Carry out various basic checks on the initial flow conditions to
ensure that the figures are realistic.
2—1 02/05/16
Step 4: Select a new fixed outlet pressure and recalculate the pressure-
dependent flows. The pressure-independent flows are assumed
to remain unchanged and it is only necessary to recalculate the
portion of the zone inflow that is influenced by the change in
pressure.
Step 5: Select a suitable time-modulated pressure distribution and
repeat the details outlined in Step 4. The selected pressure
distribution should bring the minimum pressure at the critical
point down to the minimum acceptable pressure defined for the
system. The minimum acceptable pressure may be different
from one zone to another and will depend upon several factors
including the presence of multi-storey buildings, industrial users,
fire-fighting requirements and payment levels etc.
Step 6: Select a flow-modulated pressure distribution and repeat the
various steps outlined in Step 4.
Step 7: Summarise the results and present them in a simple and easily
digestible format from which the water supplier can make an
informed decision regarding which form of pressure control (if
any) is appropriate for the system being investigated.
• Before proceeding to explain the various steps in more detail, it is necessary to

discuss two of the key parameters (the pressure exponent and the hour-day factor)
that must be clearly understood before proceeding.
2.2. PRESSURE EXPONENT (N1)

The N1 value used in Equation 1 (i.e. L1 = L0 * (P1/P0)N1) represents the power exponent
for all distribution losses in the system which is influenced by pressure. This may appear
rather confusing to those who have used the night-flow analysis model (SANFLOW)
where the burst and background leakage are dealt with separately throughout the analysis
process, with the result that there are two pressure exponents, one for burst leakage and
the other for background leakage. In the case of the pressure analysis this is not the case
and all leakage in a particular system is lumped together. The pressure exponent used in
the calculations represents a “lumped” parameter for both burst and background losses.
The value for N1 used in the PRESMAC Model will, therefore, normally vary between 0.5
(default value for bursts) and 2.5 (highest value for background leakage) with an average
or default value of 1.0. Systems with a high percentage of background leakage will tend
2—2 02/05/16
to have N1 values in excess of 1.0 while systems where the leakage is predominantly
burst leakage on iron or steel pipes will have N1 values of less than 1.0.
If possible, the value of N1 for a system should be calculated directly using measured
information obtained from a “pressure step-test” analysis undertaken during the period of
minimum night flow. The pressure step-test analysis may take between two and three
hours during which the minimum night flow would normally remain relatively constant. It
should be noted that the minimum night flow normally comprises three components:
normal night customer use, background leakage and burst leakage. Full details of the
split between the various components is provided in the SANFLOW User Guide (WRC,
1999).
The calculation of N1 is undertaken by measuring the pressure and inflow at the inlet point
and simultaneously measuring the pressure at the average zone point and critical point.
The pressure at the inlet point is then reduced in stages, allowing for the system to
stabilise before proceeding with the next pressure reduction. Normally two or three
pressure reductions can be achieved during the two- to three-hour period of constant
minimum night flow. After the consumption starts to rise due to the consumer demand, it
is not possible to continue with the pressure step-test since it is not possible to predict
accurately the influence of pressure on the consumer demand. The value of N1 is
estimated directly from the change in minimum night flow and the head losses between
the inlet and average zone points, as well as between the inlet and critical points and,
finally, between the average zone point and critical points. A typical pressure step-test
calculation is shown in Table 2.1.
Table 2.1 Calculation of N1 from pressure step-test analysis
Description Start End Pressure System Night Distribution Estimate of N1 Value

time time Inflow Use losses
Inlet AZP Critical (m3/h) Initial Stage Stage
(m) (m) (m) 1 2
Initial 01:30 64 52.0 36.0 72.0 8 64.0

Conditions
Stage 1 02:00 02:30 51 42.6 31.0 61.2 8 53.2 0.93
Stage 2 03:00 03:30 45 38 29.0 56.2 8 48.2 0.90 0.86
Stage 3 04:00 04:30 40 34 26.0 51.5 8 43.5 0.91 0.89 0.92
The example shown in Table 2.1 involves a 3-step pressure test from which 6 individual
estimates of the N1 value can be derived. The calculations of the N1 values are
presented individually for each of the six estimates as shown in Table 2.2.
The N1 values are calculated using Equation 1c as given in Section 1.2.
2—3 02/05/16
N = log (L0 / L1) / log (P0 / P1) Eq. 1(c)
Table 2.2: Calculation of the N1 values for example shown in Table 2.1
Stage Basic Head Loss Estimation of N1 N

Equation 1
For stage 1 ( 64.0 / 53.2 ) = (52.0/42.6)N1 N1 = log(1.2030) / log(1.2207) 0.93
Between stages 1 and 2 ( 53.2 / 48.2 ) = (42.6/38.0)N1 N1 = log(1.1037) / log(1.1211) 0.86
From the above analysis it is clear that an appropriate value of N1 is between 0.86 and
0.92. In this case a value of 0.9 or even 1.0 would most likely be used.
2.3. CALCULATION OF THE HOUR-DAY FACTOR

As mentioned previously, the maximum benefit to be gained from pressure reduction is at
night when the demand for water is at its lowest and system pressures are usually high.
One question asked frequently is how much water will be saved at night and how much
during the day. Obviously, the savings will depend to a large degree upon the split in
water demand between the pressure-dependent component (mainly leakage) and the
pressure-independent component (mainly consumption).
In order to provide a meaningful indication of how much water can be saved throughout
the day, a “hour-day factor” (HDF) can be used. The HDF is basically the multiplier by
which the saving during the hour of minimum night flow is multiplied to give the savings
during the full 24-hour period. If, for example, the savings are distributed evenly
throughout the day, then the HDF will be 24. In practice, however, the savings are not
usually distributed evenly, with the result that the HDF value is not 24. Values of between
10 and 40 are possible, although they normally range from a low of approximately 16 to a
high of approximately 30. A low value of 16 indicates that there is a significant difference
between the losses during the hour of MNF and the remainder of the day.
2—4 02/05/16
Example of hour-day factor calculation

In a specific zone metered area, the pressure dependant losses for the full 24-hour period
were estimated to be 1438 m3. If the pressure-dependent losses during the hour of
minimum night flow were estimated to be 64 m3 ,then the hour-day factor would be 22.47
(i.e. 1438/64).
If, however, the pressure dependent losses during the period of minimum night flow are
reduced to perhaps 30 m3 through some form of advanced pressure control, then the
hour-day factor would increase to 47.9 ( i.e. 1438/30).
The hour-day factor provides a useful indication of whether or not the estimated pressure-
dependent flows estimated in the model are realistic.
2.4. PRESSURE-DEPENDENT AND PRESSURE-INDEPENDENT FLOWS

In order to assess the impacts of pressure reduction in a zone, it is first necessary to split
the total flow entering the zone into the following two components.
• Distribution losses - influenced by pressure

• General consumption - not influenced by pressure.
How this split is achieved can be viewed as a rather subjective approach requiring some
basic assumptions. Certain variables are provided to allow the user some flexibility in this
regard and, in general, a somewhat conservative approach is usually adopted.
In summary, the various burst and background losses are assumed to be pressure-
dependent while consumption is assumed to be pressure–independent. This implies that
the three components of background leakage (connections, properties, and mains) as well
as all burst leakage are pressure-dependent while the three components of consumption
(population use, small-unmetered users and larger metered users) are all pressure-
independent.
The above assumptions are not strictly correct in that there are usually some connection
and property losses which are pressure-independent. A leaking toilet cistern for example,
which is dependent upon the water level in the cistern and not the mains pressure would
represent pressure-independent leakage. On the other hand, some of the legitimate use
which is assumed to be pressure-independent can, in fact, be pressure-dependent. Water
used for washing hands and brushing teeth, for example, is dependent upon pressure to
some extent since many people simply turn on a tap and leave it running during such
2—5 02/05/16
activities. Garden irrigation from a municipal supply is another example where domestic
consumption can be influenced by pressure since many irrigators leave a sprinkler on for
a certain duration irrespective of the quantity of water applied. As a result of such factors
domestic consumption is known to decrease to some degree as the pressure drops. In
view of these considerations and to err on the conservative side, the flow which is
assumed to be independent of pressure is considered to be the following:
Components of flow independent of pressure
Consumption
• Normal domestic use The component influenced by pressure is NOT
• Small non-domestic users taken into account at this stage to provide a
• Large-metered users conservative estimate of the potential savings
Background Losses
• A portion of the connection losses
(default value 0.5 l/conn.h = ± 15% of 3 l/conn.h)
• A portion of the property losses

(default value 0.5 l/prop.h = ± 50% of 1.0 l/prop.h)
In order to split the total flow into the pressure-dependent and pressure-independent
components for each hour of the day, the following approach is adopted.
• Estimate the pressure-independent component during the hour of minimum night flow
based on the BABE methodology.
• From the above, estimate the pressure-dependant component at the same hour from
the MNF
• Having established the pressure-dependent losses during the hour of MNF for which
the average zone pressure is known, the pressure-dependent component for each of
the other 23 hours can be estimated using Equation 1 as discussed in Section X.X.
• Having established the pressure-dependent losses, the pressure independent
component can be estimated by simply subtracting the pressure-dependent losses
from the total zone inflow for each hour of the day.
The process of splitting the total flow into the two components may initially seem
confusing. It is important to note that it is only at the point of minimum night flow that the
2—6 02/05/16
consumptive use can be estimated from first principles. At the point of minimum night flow
the consumptive use can be estimated from the population and average use per person
per hour (usually 6% of population times 10 litres per hour). The water used by other
small consumers such as garages, etc as well as large consumers is also taken into
account in order to estimate the actual consumption. Having established the
consumption, the difference between it and the minimum night flow then provides an
indication of the quantity of water lost through burst and background leakage. It is
important to note that for the purpose of the PRESMAC Model, the burst and background
leakage are treated as one component and that the N1 pressure exponent used in the
model is a “lumped parameter”. This is a subtle difference from the SANFLOW night-flow
analysis model where burst and background leakage are treated separately with the result
that two N1 exponents are used.
The outputs from this stage of the analysis are two sets of 24-hour values representing
the pressure-dependent and pressure independent flows.
To explain the calculation, a simple example is provided in Table 2.4 using an initial
pressure-independent flow of 8m 3/h (as estimated below in Table 2.3). The calculations
are carried out as follows:
Table 2.3: Estimation of pressure-independent flow at MNF (Item 1 in Table 2.6)
Description Basis of calculation Calculated

flow
3
Losses per connection 641 connections * 0.5 l/con.h = 0.32 m /h
3
Losses per property 2210 properties * 0.5 l/property.h = 1.11 m /h
3
Residential night use 9945 people * 6% * 10 l/h = 5.97 m /h
3
Non-residential (unmetered) 7 café’s @ 50 l/h = 0.35 m /h
3
Non-residential (metered) = 0.26 m /h
Total pressure independent flow at MNF 8.00 m3/h
The pressure dependent flow can now be estimated as shown in Table 2.4 for the hour of
MNF.
Table 2.4: Estimation of pressure-dependent flow at MNF (item 2 in Table 2.6)
Description Basis of calculation Calculat

ed flow
Total metered flow – pressure- 72.00 m 3/h – 8 m 3/h 64 m 3/h
independent flow
2—7 02/05/16
Having established both the pressure-dependent and pressure-independent flows during

the hour of MNF, it is then possible to estimate the pressure-dependent losses for all other
hours. An example of the calculation for the first hour (0 to 1) is provided in Table 2.5.
Table 2.5: Estimation of pressure-dependent flow at hour 0 – 1(item 3 in Table 2.6)
Description Basis of calculation Calculated flow

Use equation 1b where L1 = 64 m 3/h x (61 m / 64 m)1.0 61.53 m 3 / h
L1 = L0 * (P1/P0)N1
The calculation of the pressure-dependent flow is repeated for each remaining hour until
the pressure-dependant losses have been calculated for the full 24-hour period
(calculations 5 to 25 in Table 2.6)
Table 2.6 Example to demonstrate the calculation of pressure-dependent and

pressure-independent flow
Hour Inlet Total Pressure- Order of Pressure- Order of

Pressure Inflow Dependent Calculation Independent Calculation
3 3
(m) (m /h) Flow Flow (m /h)
3
(m /h)
0–1 61 82.80 61.53 3 21.27 26
1–2 63 75.60 62.76 4 12.84 27
2–3 64 72.00 64.00 5 8.00 28
3 – 4* 64 72.00 64.00 2 8.00 1
4–5 64 72.00 64.00 6 8.00 29
5–6 64 75.60 64.00 7 11.60 30
6–7 61 93.60 62.76 8 30.84 31
7–8 60 108.00 61.53 9 46.47 32
8–9 57 111.60 57.84 10 53.76 33
9 – 10 57 115.20 57.84 11 57.36 34
10 – 11 57 118.80 57.84 12 60.96 35
11 – 12 57 115.20 57.84 13 57.36 36
12 – 13 57 111.60 57.84 14 53.76 37
13 – 14 56 115.20 56.61 15 58.59 38
14 – 15 56 111.60 56.61 16 54.99 39
15 – 16 57 111.60 57.84 17 53.76 40
16 – 17 56 104.40 56.61 18 47.79 41
17 – 18 57 104.40 57.84 19 46.56 42
18 – 19 57 108.00 57.84 20 50.16 43
19 – 20 58 108.00 59.07 21 48.93 44
20 – 21 58 104.40 59.07 22 45.33 45
21 – 22 59 100.80 60.30 23 40.50 46
22 – 23 60 100.80 61.53 21 39.27 47
23 – 24 60 97.20 61.53 25 35.67 48
* hour of minimum night flow
2—8 02/05/16
Having established the pressure-dependent flows for each hour of the 24-hour period, it is
a very simple calculation to estimate the remaining pressure-independent flows for each
hour by simply subtracting the pressure-dependent flows from the zone inflow. The
results are indicated in Table 2.6 (calculations 26 to 48)
The breakdown between pressure-dependent and pressure-independent flows for each

hour in the 24-hour period has now been established as shown in Table 2.6.
2.5. ESTIMATION OF NETWORK FRICTION (“K”) FAC TORS

Having established the split between pressure-dependent and pressure-independent
flows, the next step is to estimate the friction factors (K factors) for each hour of the day.
The hourly “K” factors effectively represent the friction loss for the whole distribution
system for a specific hour of the day. Two sets of “K” factors are used throughout the
analyses, namely, those representing the friction losses from the inlet to the Avarage
Zone Point (AZP) and, secondly, those representing the losses from the inlet to the critical
point. The underlying assumption is that, for each hour, it is a “lumped” factor for the
system as a whole and that although the factors will vary from hour to hour, they will
remain approximately the same for each day. In other words, the same “K” factor will be
used for hour 12 to 13 on day 1 as for hour 12 to 13 on day 2, day 3, day 4, etc.
This is a rather brood assumption and can best be understood by using an analogy of a
road traffic system. If the water distribution system is compared to the traffic flow pattern
of a similar sized town, the use of the “K” factor can be conceptualised. In a traffic
system, the traffic flow patterns tend to be similar from day-to-day although they may vary
considerably from hour-to-hour throughout the day depending upon when the peak hour
traffic into and out of the main centre occurs. The pattern of water flow is assumed to vary
much in the same manner, hence, the use of a separate K factor for each hour of the day.
This simplistic approach has been found to provide realistic results in most instances,
particularly at the critical point. The “K” values at the AZP often appear to fluctuate
considerably. However, this is not considered to be a problem since the end results are
considered within the acceptable tolerances for this basic approach.
The “K” factors are calculated in the following manner:
From equation 2
H L = K x Q2 or K = HL /Q2
2—9 02/05/16
Where HL = head loss (m)

Q = flow (m3/h)
K = friction factor/coefficient (m-5h2)
Two “K” factors for each hour of the day are calculated using the above equation where
the HL term is the difference in pressure between the inlet point and the AZP as well as
between the inlet point and the critical point. Details of the calculation of the “K” factors
for the target and critical points are given in Table 2.7.
The calculation of the “K” factors for hour 0 – 1 is provided in Table 2.7.
Table 3.7: K factors for AZP and critical points (items 2 and 3 in Table 2.8)
Description Basis of calculation K factor

2
K factor at the AZP where : K = 11 / (82.8) 0.00160
∆H = 11m and 4
Q = 82.8 m 3/h
K factor at the critical point where : K = 31 / (82.8) 2 0.00452
∆H = 31 m and 2
Q = 82.8 m 3/h
The full set of K-factors is provided in Table 2.8 for a specific data set.
As can be seen in Table 2.8, the “K” factors for the critical point appear reasonably
consistent as would be expected. The “K” factors for the AZP, however, seem to vary
considerably and tend to look unrealistic in some cases. This is not a problem and should
also be expected, due to the fact that the position of the AZP is often not known and has
been approximated.
Having established the split in the pressure-dependent and pressure-independent flows

as well as the K-factors for each hour of the 24-hour period, it is now possible to test the
impact of various possible pressure management options on the leakage from the zone.
The first option to be considered is the installation of a standard pressure-reducing valve
(PRV). In cases where a PRV is already in operation, it is possible to test the influence of
changing the fixed pressure level.
2—10 02/05/16
Table 2.8: Estimation of K-values at AZP and critical point

∆ H Between
Pressure (m) Flow K factors for
Hour inlet
m3/h
Inlet (m) AZP (m) Critical (m) AZP Critical AZP Critical
0–1 61 50 30 11 31 82.8 0.001604 0.004522
1–2 63 51 34 12 29 75.6 0.00210 0.005074
2–3 64 52 36 12 28 72.0 0.002315 0.005401
3 – 4* 64 52 36 12 28 72.0 0.002315 0.005401
4–5 64 52 36 12 28 72.0 0.002315 0.005401
5–6 64 52 36 12 28 75.6 0.002100 0.004899
6–7 61 51 25 10 36 93.6 0.001141 0.004109
7–8 60 50 18 10 42 108.0 0.000857 0.003601
8–9 57 47 17 10 40 111.6 0.000803 0.003212
9 – 10 57 47 17 10 40 115.2 0.000754 0.003014
10 – 11 57 47 17 10 40 118.8 0.000709 0.002834
11 – 12 57 47 17 10 40 115.2 0.000754 0.003014
12 – 13 57 47 17 10 40 111.6 0.000803 0.003212
13 – 14 56 46 16 10 40 115.2 0.000754 0.003014
14 – 15 56 46 16 10 40 111.6 0.000803 0.003212
15 – 16 57 47 17 10 40 111.6 0.000803 0.003212
16 – 17 56 46 19 10 37 104.4 0.000917 0.003395
17 – 18 57 47 19 10 38 104.4 0.000917 0.003486
18 – 19 57 47 20 10 37 108.0 0.000857 0.003172
19 – 20 58 48 20 10 38 108.0 0.000857 0.003258
20 – 21 58 48 22 10 36 104.4 0.000917 0.003303
21 – 22 59 49 25 10 34 100.8 0.000984 0.003346
22 – 23 60 50 26 10 34 100.8 0.000984. 0.003346
23 – 24 60 50 28 10 32 97.2 0.001058 0.003387
2.6. ANALYSIS OF FIXED-OUTLET PRV

The fixed- outlet PRV option is the simplest and most basic of the various pressure
management options that can be considered. It involves installing a PRV at the inlet to a
zone which will ensure that the pressure entering the zone remains at a constant level
throughout the day. (refer to Fig. 1.1 in Section 1.1)
The objective of installing a PRV at the inlet point is to lower the pressure at the target or
critical point until it meets the minimum level of service.
If leakage and head losses were uniform and linear in nature it would be a very simple
operation to estimate how much the inlet pressure should be reduced in order to provide
exactly the minimum pressure allowed at the critical point during the period of peak
demand. Unfortunately, life is not quite as simple and a pressure drop of 10 m or more at
the inlet point is often required to achieve a drop in pressure of only 5 m at the critical
2—11 02/05/16
point. This is an important point that is often overlooked when considering the possibility
of pressure reduction in a zone. The distribution engineer often indicates that the excess
pressure at the critical point is perhaps only 3 or 4 m and, as such, he/she is of the
opinion that there is little, if anything, to be gained from pressure reduction. If it is shown ,
however, that the inlet pressure can be dropped by more than 10 m to achieve the drop of
only 3m at the critical point, then there are significant gains to be made.
In order to model this process in the PRESMAC Model, a very simple iterative approach is
adopted in which the following methodology is employed.
Step 1
• An initial estimate is made of the fixed outlet pressure that will yield the minimum
acceptable pressure at the critical point during the hour of minimum pressure (i.e. peak
demand period). For example, if the minimum pressure at the critical point is 25 m and
the minimum level of service is 20 m, then the inlet pressure can be reduced by at least
5 m. This estimate of 5m is then subtracted from the actual pressure at the inlet point
during the critical hour and then applied for the full 24-hour period representing a static
outlet pressure from the PRV.
Step 2
• Having set a constant pressure for the full 24-hour period, it is then necessary to re-
calculate the new pressures at the AZP and critical point. This is an iterative procedure
and will be described in detail later in this section.
Step 3
• The new pressure at the critical point is checked against the minimum pressure
requirement and, if possible, the inlet pressure is reduced further. This procedure is
repeated until the pressure at the critical point is in line with the desired minimum
pressure.
The above procedure is best explained through the use of a simple example.
Example
In this example, the minimum pressure at the critical point is measured to be 16 m
between hours 13 and 15 during which the inlet pressure is measured to be 56 m. In this
case the minimum level of service is only 10m and the inlet pressure can be reduced by at
least 6 m.
2—12 02/05/16
Step 1: Initial estimate of new fixed-outlet pressure

The initial estimate for the fixed outlet pressure will therefore be set at 50 m (59 m – 6 m).
In practice we would probably start with a reduction of 10 m or 12 m. However, to
demonstrate the procedures, an initial reduction of 6m will be used in the knowledge that
the exercise will have to be repeated several times until a pressure of 10 m is obtained at
the critical point.
The new inlet pressure to 50 m is now used for each hour as indicated in Table 3.9.
Step 2: Calculate the resulting pressures at the AZP and critical point
Using the new fixed-outlet pressure, the resulting pressures at both the AZP and critical
point are for the first hour and then for each of the 23 remaining hours. At this iterative
stage in the analysis, it is only necessary to analyse the hour of peak demand when the
pressure at the critical point will be at its minimum. Since the whole procedure is
undertaken almost instantly within the PRESMAC Model, the calculation is carried out for
all 24 hours.
The calculation is undertaken in an iterative manner in which the “K” values calculated
earlier are used. The procedure for the first hour at the AZP is as follows:
Step 2a
• For the first hour the initial pressure-dependent losses were estimated to be 61.53 m3 for
an inlet pressure of 61 m. The inlet pressure has been reduced to 50 m (i.e. a drop of 11
m) and so an 11m drop at the AZP to 39 m (i.e. 50 m – 11 m) is used as the first estimate.
Step 2b
• The pressure-dependent losses are re-calculated for the new pressure at the AZP from
Equation 1. i.e.
L1 = L0 x (P1/P0)N1
where N1 in this example is assumed to be 1.0 which produces a new estimate of the
pressure-dependent losses as 48 m3/h (i.e. 61.53 * (39/50)). This calculation indicates
that the pressure-dependent losses will reduce from 61.53 m3/h to 48 m3/h during the first
hour if the inlet pressure is reduced from 61 m to 50 m. This clearly demonstrates how
important pressure can be with regard to system leakage.
2—13 02/05/16
Step 2c
• Using the new estimate of pressure-dependent losses of 48 m3/h, the previously
calculated pressure-independent losses of 21.27 m3/h are added to give a new estimate
of the total system inflow. It should be remembered that the pressure independent
losses/use do not change with the decrease in the zone inlet pressure since they are
theoretically independent of pressure. The new zone inflow is estimated to be 69.27 m3/h
(i.e. 48 + 21.27) compared to the original value of 82.8 m3/h.
Step 2d
• Using the new estimate of the total zone inflow of 69.27 m3/h, the appropriate K value
from Table 3.3 (i.e. 0.001604) is used together with equation 2 to re-estimate the head
loss between the inlet point and the AZP.
i.e. new head loss = 0.001604 * 69.272 = 7.7 m
Step 2e
• The revised head loss of 7.7 m is now compared to the initial estimate of 11 m if the
difference between the two estimates is significant. The inlet pressure is altered
accordingly and the whole calculation is repeated. In this example it is clear that the head
loss (11 m) has been overestimated and the new lower value is now used in the second
iteration. As a result, a new estimate of the pressure at the AZP is made, using 7.7 m and
not the previous estimate of 11 m. It should be remembered that the purpose of this
calculation is to establish the new pressure at the AZP for the new inlet pressure of 50 m
(previously 61.3 m).
The iterative procedure is shown in Table 2.9.
Table 2.9: Iterative approach used to calculate the new pressure at the AZP
1 2 3 4 5 6
Iteration Estimate Estimated Pressure- Total zone Calculate new New estimate
No. Pressure at Pressure- independent inflow (2 + 3) head loss between of AZP
AZP dependent losses (m3/h) inlet and AZP
(m3/h) Pressure
(m) losses (m3/h) (m) (m)
1 39.0 48.0 21.27 69.27 7.7 42.3
2 42.3 52.0 21.27 73.32 8.6 41.3

3 41.3 50.9 21.27 72.19 8.36 41.6
4 41.6 51.2 21.27 72.57 8.43 41.6
72.44
2—14 02/05/16
From Table 2.9 it can be seen that the pressure at the AZP stabilises at 41.6m after 4
iterations.
The same calculation is then undertaken for the remaining 23 hours at the AZP.
Having calculated the new zone inflow for each hour as well as the pressure at the AZP,
the next step is to estimate the new hourly pressures at the critical point. In this case it is
not necessary to carry out an iterative approach, but instead, make use of the zone inflow
and the previously calculated K values which are used in Equation 2.
i.e. HL = K x Q2
In hour 1 the new Q value has been calculated to be 72.44 m3/h (see Table 2.10) and the
K value 0.004522. Using these values the new head loss is calculated from
HL = 0.004522 x 72.442 = 23.73 m
The new pressure at the critical point for hour 1 is estimated to be:
50 m – 23.73 m = 26.3 m
This simple calculation is repeated for each hour until all pressures have been estimated.
The resultant pressures at both the AZP and critical points are provided in Table 2.10.
Step 3: Check new pressure at critical point and adjust if necessary

From Table 2.10 it can be seen that the drop in pressure to 50 m has resulted in a drop in
pressure at the critical point to a minimum value of 14.1 m. In other words, a 6 m drop in
inlet pressure has resulted in only a 1.8 m drop in pressure at the critical point. From this
result it is clear that the inlet pressure can be lowered significantly more than 6 m to obtain
a minimum pressure of 10 m at the critical point.
The process is repeated several times until the desired minimum pressure of 10 m is
obtained. The results from the iterative process are shown in Table 2.11.
2—15 02/05/16
Table 2.10 Revised pressures at AZP and critical points for fixed outlet PRV.
Pressure at AZP (m) Pressure at Critical Point (m)

Hour
Original If Inlet = 50 m Original If Inlet = 50 m
0–1 50 41.6 30 26.3

1–2 51 41.5 34 29.3
2–3 52 41.8 36 30.9
3–4 52 41.8 36 30.9
4–5 52 41.8 36 30.9
5–6 52 41.7 36 30.6
6–7 51 42.2 25 21.9
7–8 50 41.8 18 15.5
8–9 47 41.2 17 14.9
9 – 10 47 41.2 17 14.8
10 – 11 47 41.2 17 14.7
11 – 12 47 41.2 17 14.8
12 – 13 47 41.2 17 14.9
13 – 14 46 41.0 16 14.1
14 – 15 47 41.1 16 14.2
15 – 16 47 41.2 17 14.9
16 – 17 46 41.1 19 17.1
17 – 18 47 41.3 19 16.9
18 – 19 47 41.3 20 17.7
19 – 20 48 41.4 20 17.5
20 – 21 48 41.5 22 19.3
21 – 22 49 41.7 25 21.8
22 – 23 50 41.9 26 22.4
23 – 24 50 41.9 28 24.2
Table 2.11 Derivation of new inlet pressure to provide critical pressure of 10m
Iteration PRV Pressure (m) Minimum pressure at

critical point (m)
Initial n/a 16.0

1 50 14.1
2 46 12.8
3 43 11.7
4 39 10.2
5 38.5 10.0
From the results shown in Table 2.11 it can be seen that a PRV pressure of 38.5 m can
be used to produce a minimum pressure of ±10m at the critical point.
2—16 02/05/16
The results produced from the program are now used to estimate the total savings that will
be achieved through the use of the new PRV set at 38.5 m. Details of the savings are
provided in Table 2.12.
Table 2.12 Savings achieved through the use of PRV set at 38.5 m
Zone Inflow (m3/h) Saving achieved

Hour
Initial For PRV at 38.5 m m3/h
0–1 82.8 61.2 21.6

1–2 75.6 53.0 22.6
2–3 72.0 48.7 23.3
3–4 72.0 48.7 23.3
4–5 72.0 48.7 23.3
5–6 75.6 52.0 23.6
6–7 93.6 71.1 22.5
7–8 108.0 86.0 22.0
8–9 111.6 92.7 18.9
9 – 10 115.2 96.2 19.0
10 – 11 118.8 99.7 19.1
11 – 12 115.2 96.2 19.0
12 – 13 111.6 92.7 18.9
13 – 14 115.2 97.2 18.0
14 – 15 111.6 93.7 17.9
15 – 16 111.6 92.7 18.9
16 – 17 104.4 86.7 17.7
17 – 18 104.4 85.7 18.7
18 – 19 108.0 89.2 18.8
19 – 20 108.0 88.1 19.9
20 – 21 104.4 84.6 19.8
21 – 22 100.8 80.1 20.7
22 – 23 100.8 79.1 21.7
23 – 24 97.2 75.6 21.6
Daily Totals 2390.4 1899.6 490.8
It should be noted that the savings indicated in Table 2.12 are conservative and, in reality,
are likely to be higher since no allowance has been made for a reduction in consumptive
use or the reduction in leakage due to the reduced number of burst pipes. Although the
consumptive use is initially assumed to be independent of pressure, this is not strictly
accurate since some of the consumptive flow will be influenced by pressure. Recent
studies by John May (personal communication) indicate that toilet cisterns can use up to
2—17 02/05/16
an additional 10% of normal flush volume if connected directly to mains pressure. This is
due to the additional flow occurring as the cistern is emptying and the valve is fully open.
The results provided also exclude any savings from reduced bursts which are usually
calculated separately. The savings indicated in the PRESMAC Model only refer to the
direct savings from reduced system input for the system in its current condition.
2.7. ANALYSIS OF TIME-MODULATED CONTROLLER

By using a fixed-outlet PRV, it is possible to reduce the system pressure to a level
whereby the critical point receives the minimum level of service (i.e. 10 m in the previous
example) during the period of maximum demand.
This is normally the point at which the engineers and system designers feel satisfied that
they have completed their duty. In reality, however, the pressure at the critical point will
be considerably higher than the minimum level of service for most of the time when the
system demand is lower. In some respects this may be acceptable since it implies that
everyone in the system is receiving at least the minimum pressure throughout the day.
From a leakage viewpoint, however, it also implies that the leakage rates are likely to be
considerably higher than they need to be during all times, except the one or two hours
when the critical point is receiving the minimum allowable pressure as discussed in
Section 1.2.
A time-modulated PRV consists of an electronic timing device that will drop the inlet
pressure to a lower setting during certain times of the day or night. It is the simplest and
least expensive form of advanced pressure control and enables the engineer to drop the
pressure during periods of low demand.
Various time-modulated controllers are currently available which allow the engineer to
select two pressures (high and low) as well as four switching periods. This enables the
pressure to be switched from high to low up to four times per day. In many instances, it is
found that only two switching periods are required and not the maximum of four available
through the controller. Normally it is only necessary to drop the pressure from perhaps
8pm to 6am and leave it at the higher setting for the rest of the day.
It should be noted that certain time-modulated controllers can also be used in association
with a flow meter to offer a crude form of flow-modulated control. In the flow-modulating
mode such controllers can switch from low to high pressure if the flow sensor indicates a
2—18 02/05/16
flow higher than some pre-defined limit. The use of a time-modulated controller operating
in flow mode is not accommodated in the PRESMAC Model.
The PRESMAC model allows the user to simulate a time-modulated controller in order to
assess the potential savings that can be achieved through its use. Although this option
may not provide the higher savings that can often be achieved through the use of a flow-
modulated controller, it will provide a realistic estimate of the savings that can be achieved
through the use of the simpler and less expensive controller.
In order to use the time-modulating option, the user of the program simply selects a period
(or up to two periods) during which the pressure is lowered from the high value to a lower
value. For example, the user may wish to consider reducing the pressure at a fixed outlet
PRV from 50m to 30m during the low demand period from 10pm to 5am as shown in
Fig. 2.1.
Figure 17.CDR
Inlet pressure
50m
30m
0 2 4 6 8 10 12 14 16 18 20 22 24
Time (hours)
Figure 2.1 Example of a (two-point) time -modulated pressure profile
The time modulated option is analysed in exactly the same manner as the fixed-outlet
PRV option, whereby the new pressures at the Average Zone Point (AZP) are first
calculated in an iterative approach using the K factors. Having established the new
pressures at the AZP and the corresponding zone inflows, the pressures at the critical
point are estimated. The time-pressure profile can then be adjusted if required by either
changing the times for switching from high to low pressure or by changing the pressure
settings. As indicated previously, up to four switching levels can be used which will allow
two blocks of low pressure to be modelled. It must be remembered that only two pressure
settings are permitted – i.e. a high pressure and a low pressure.
2—19 02/05/16
3. USING PRESMAC
3.1. HARDWARE AND SOFTWARE REQUIREMENTS

In order to run the PRESMAC Model the user requires a standard personal computer with
the Windows 98 (or above) operating system. The model has been developed for use on
colour screens with a resolution of 800 by 600 or better.
3.2. INSTALLING PRESMAC

PRESMAC is supplied on a single disc which contains two files namely PRESMAC.EXE
and PRESMAC.PRU. The PRESMAC.EXE file is the main executable which can be run
by simply double clicking on the file from any standard file management program. The
PRESMAC.PRU file is a sample data file providing the user with a data template from
which other data files can be generated.
To load and run PRESMAC, the user should create a directory (called PRESMAC for
example) and copy both files from the disc into the directory. The executable can then be
run without any additional software. It should be noted that the model is not accompanied
by an installation shield with the result that the user must create his/her own icons and
items on the start-up menu if desired.
3.3. DATA REQUIREMENTS

In order to use PRESMAC the user must first obtain the relevant data for the area to be
examined. Unfortunately, the data collection can often be time-consuming and
problematic due to the fact that a certain amount of logging information is required. In
general, the data required to operate the model are as follows:
General Zone Information

• Length of mains;
• Number of connections;
• Number of properties;
• Population;
• Condition of network (mains, connections, properties)
• Type of properties (informal, affluent etc)
• Estimate of illegal connections
Logged Flow and Pressure

• 24-hour profile of zone inflow;
• 24-hour profile of pressure at inlet point
3—1 02/05/16
• 24-hour profile of pressure at critical point;

• 24-hour profile of pressure at Average Zone Point.
In order to facilitate the data capture and reporting of data for each zone, a data input form
has been prepared and is provided in Appendix B.
The logging information should ideally refer to a relatively extreme period when the
system demands are at or near their highest. In some instances, it may be appropriate to
consider two events, one during normal demand periods and the other during extreme
demand periods. This is particularly relevant for areas experiencing large fluctuations in
demand as is often the case in holiday locations or areas where garden irrigation during
the hot summer periods is significant. Due to the flexibility of the advanced pressure
control devices, it is relatively simple to define two (or more) operating procedures for the
different periods of the year and the equipment can then be adjusted at the start of the
winter and summer periods or as appropriate.
3.4. RUNNING PRESMAC

As mentioned previously, the PRESMAC Model is opened by double clicking on the
PRESMAC.EXE file. This opens the PRESMAC Model which is a self-contained DELPHI
Program requiring no additional software (e.g. EXCEL). After opening the model the user
should see the standard front page as shown in Fig. 3.1. This page remains on the
screen for several seconds to give the user sufficient time to read the version number, etc.
Figure 3.1: Opening banner from PRESMAC
3—2 02/05/16
After a few seconds, the main screen appears which allows the user to start running the
model. Generally, the first action to be taken will be to open an existing data file. If the
user has not created any data files previously, the sample data file PRESMAC.PRU
should be opened. The main screen is shown in Fig. 3.2.
Figure 3.2 : Main menu in PRESMAC
As can be seen in Fig. 3.2 there are five items on the main menu bar namely:
• Data Input
• Current Situation
• Fixed-Outlet PRV
• Time-Modulated PRV
• Summary
Each of the above items is discussed separately in the remainder of Section 3.
3.5. DATA INPUT

The Data_Input form provides various screens to allow the user to supply details of the
system as well as the information required to analyse the potential for pressure
management. The information required includes the items indicated in Section 3.3 plus
3—3 02/05/16
other details such as the zone name and reference number, etc. When the Data_Input
tab is selected, the user is provided with four sub-forms (see Fig. 3.2) to choose from
namely:
• General_Info
• Default_Parameters
• Flow_and_Pressure
• N1_Calculations
3.5.1. General_Info
The General_Info form incorporates five blocks for data input namely:
• Zone Reference Information;
• Contact Details;
• Elevations of inlet, AZP and critical points;
• Hour number for minimum and maximum flow;
• Property information.
The items are self-explanatory and need no further description. The boxes are colour-
coded, where red represents information that must be supplied by the user while the
property information is given in green indicating that it is used elsewhere in the model.
The hour number for minimum and maximum flow is shown in black which indicates that it
is calculated directly from the data supplied by the user. It is possible to adjust the values
calculated by the model in case there are several consecutive hours with the same
minimum or maximum flow.
3.5.2. Default Parameters

If the Default_Parameters tab is selected, the form shown in Fig. 3.3 will be displayed.
This form allows the user to select and input various parameters that are used in the
subsequent calculations. As before, the form is split into several blocks which contain
related information. The following data blocks are included:
• System Leakage Parameters;
• Unit usage data during periods of Minimum Night Flow;
• Daily non-residential consumption;
• N1 Value;
• Night Consumption Calculations.
3—4 02/05/16
Figure 3.3: The Default_Parameters Form (from the Data_Input Form)
System Leakage Parameters

The user is required to supply information concerning the leakage parameters used to
estimate the background leakage. Again it should be noted that all parameters are related
to the standard pressure of 50 m. The information required is basically the same as used
in the SANFLOW Model where the background leakage parameters are used to calculate
the background leakage from the mains, connections and properties. In this case,
however, the information is supplied in a slightly different form in that the parameters are
split into pressure-dependent and pressure-independent components. The user is able to
select a realistic parameter value and then indicate the split so that the background
leakage can be analysed in a more realistic manner. It should be noted that all of the
mains leakage is considered to be pressure-dependent, with the result that the user is not
allowed to split the parameter value (40 l/km.h in the example). With regard to the
connection leakage, a small portion is usually assumed to be pressure-independent and in
the example 0.5 l/conn.h from the total of 3.0 l/conn.h is considered to be pressure-
independent. Similarly, with the property leakage, 50% of the leakage is assumed to be
pressure-independent in Fig. 3.3, which could be used to take leakage from a toilet cistern
into account for example.
3—5 02/05/16
Only the green figures can be changed by the user while the blue figures have been
calculated and cannot be changed.
Unit Usage Data During Period of MNF

The information supplied by the user in this portion of the form is identical to that used in
the SANFLOW Model and is relatively simple and straightforward.
• The Cistern Capacity is the capacity of the toilet cistern and is usually 10 litres or
sometimes 6 litres.
• The % of Population Active represents the percentage of the population who will
flush the toilet during the hour of minimum night flow. The standard percentage used
in most cases is 6%, although this may vary from area to area. In areas serviced by
standpipes the value will be zero.
• The Non-Residential Night Use is the unit use per hour for the non-residential units
as input on the Data_Input form under the Property_Information block. Normally it will
refer to a unit user for garages or all-night cafes, etc and a value of between
20 l/prop/h to 50 l/prop/h will be appropriate.
• The Exceptional Night Use refers to the water used by any large users where the
actual meter readings are used to establish the water use during the period of
minimum night flow.
Daily Non-Residential Consumption

The daily non-residential consumption is used in the model to estimate the average use
per head of population per day and is not used in any other calculations. The user must
try to estimate the total use per day by industry and commerce in order to estimate how
much water is used for normal population consumption.
3.5.3. Flow and Pressure

The flow and pressure form is used to input the four 24-hour flow and pressure profiles
which form the basis for the pressure management calculation. When this form is opened
the user should see the details shown in Fig. 3.4.
3—6 02/05/16
Figure 3.4 : Details of the Flow and Pressure Form
As can be seen from Fig. 3.4, five columns are given of which four are in red and one in
blue. The blue column is a calculated column and depends upon the units selected for
the zone inflow. If the user selects m3/h as the unit for input, then the column for m 3/h will
be in red indicating that the user must supply this data. The column in l/second will then
appear in blue since it is calculated directly from the figures given in m3/h. The light blue
and yellow shaded areas refer to the hours of minimum and maximum flow respectively.
3.5.4. N1 Calculations
One of the most important factors influencing leakage is pressure. Considerable work has
been undertaken over the past 10 years in many parts of the world to establish how
leakage from a water distribution system reacts to pressure.
It is generally accepted that flow from a hole in a pipe will react to pressure in accordance
with normal hydraulic theory that indicates a square root power relationship between flow
and pressure.
3—7 02/05/16
FlowP2 = FlowP1 x (P1/P2) N1
where:
P1 = Pressure 1 (m)
P2 = Pressure 2 (m)
Flow P1 = Flow at pressure P1 (m3/h)
Flow P2 = Flow at pressure P1 (m3/h)
N1 = power exponent.
This implies that if pressure doubles, the flow will increase by a factor of 1.4 (i.e. 2 0.5).
This has been tested and found to be realistic, irrespective of whether the pipe is above
ground or buried. The problem arises because in many systems the leakage has been
found to react by a factor greater than 1.4. This has caused considerable debate and
confusion especially when trying to establish the likely savings through pressure-reduction
measures.
Although there are still various opinions concerning the explanation for the larger than
expected influences of pressure on leakage in many systems, at least one plausible
theory has been suggested. In 1994, John May (May, 1994) in the UK first suggested the
possibility of fixed area and variable area discharges (FIVAD). He carried our
considerable research on this topic and has found that systems will react differently to
pressure depending upon the type of leak being considered. If the leak is a corrosion hole
for example, the size of the opening will remain fixed as the pressure in the system
changes on a daily cycle. In such cases, the water lost from the hole will follow the
general square root principle as outlined above. This type of leak is referred to as a fixed
area leak.
If, however, the leak is due to a leaking joint, the size of the opening may, in fact, increase
as the pressure increases due to the opening and closing of the joint with the changing
pressure. In such cases the flow of water will increase by much more than the fixed area
leak. Research suggests that in such cases, a power exponent of 1.5 should be used
instead of the 0.5 used for the fixed area cases. This suggests that if the pressure
doubles, the leakage will increase by a factor of 2.83 (i.e 2 1.5).
In the case of longitudinal leaks, the area of leak may increase both in width as well as in
length as is often the case with plastic pipes. In such cases the power exponent can
increase to 2.5. In other words, if the pressure doubles, the flow through the leak will
increase by a factor of 5.6 (i.e. 2 2.5).
3—8 02/05/16
The problem faced by the water distribution engineer is to decide what factor should be
used when estimating the influence of pressure on leakage flow. In general, it is
recommended that a power exponent of 0.5 should be used for all burst flows since a
burst pipe is usually a fixed area discharge. In the case of the background losses,
however, the leaks are likely to be variable area discharges in which case a larger power
exponent should be used. A power exponent of 1.5 is usually used for the background
losses, which is considered to represent a collection of leaks that have factors of between
0.5 and 2.5. If all of the pipe work is known to be plastic, a higher value may be
appropriate and conversely, if the pipes are made from cast-iron, a lower value (eg 1.0)
should be used. Full details of the FIVAD principles are provided in Appendix B.
The N1 value used in PRESMAC is neither the burst or background N1 value as

mentioned above. In this case, the N1 value is a lumped parameter representing the
variation of leakage to pressure for the system as a whole where burst and background
leakage is considered together. From various tests undertaken around the world, it
appears that the average N1 value for a system is in the order of 1.15 and for the purpose
of the pressure management model, it is generally appropriate to adopt a conservative
value of 1.0.
In some cases it is possible to establish the true N1 value for a system through a series of
pressure “step-tests” which should be carried out during the period of minimum night
flows. To carry out the tests, the user drops the pressure by 10 or 20 m and allows the
system to settle down to the lower pressure. The minimum night flow is continually
monitored as are the pressures at the inlet, average zone point and critical point. From
the information recorded, it is possible to establish the N1 value several times from which
an average value can then be selected. A typical analysis is shown in Fig. 3.5, where it
can be seen that an N1 value of 0.92 would be appropriate for the system being analysed.
In this case, the user can manually type in the new value in the Default_Parameters form
or click the “Update” button which will automatically copy the new value of N1 as
calculated on the sheet to the correct location on the Default_Parameters form.
If there is no information on the N1 value and there is no time to carry out a pressure
“step-test”, then a value of 1.0 is normally accepted as realistic for a system. If it is known
that the system is all metal pipes, a lower value may be used and if the pipes are all
plastic, a higher value should be selected.
3.6. CURRENT SITUATION
The Current_Situation form effectively provides details for the zone before any form of
pressure management is implemented. The form basically provides the data supplied by
the user and, in addition, provides a split between pressure-dependent and pressure-
3—9 02/05/16
independent flow. The information is available in tabular form as shown in Fig. 3.6 and
also in graphical form as shown in Fig. 3.7.
Figure 3.5 : Example of a Pressure Step-Test Analysis
Figure 3.6: Tabulated Information from the Current_Situation Form
3—10 02/05/16
Figure 3.7: Graphical Output from the Current_Situation Form
It should be noted that the user can customise the graph to his/her own requirements by
selecting the “Graph Properties” button. This, in turn, will provide a new form where the
various lines shown on the graphs can be switched on or off. In addition, the scales for
the pressure and flow can be defined which will then overwrite the default scales. This
option is useful if the graphs from the different options are to be compared with each
other. The graph options are indicated in Fig. 3.8.
Figure 3.8: Graph options available
3—11 02/05/16
It should be noted that the graph options available for the “Current_Situation” exclude any
head loss items although these are available on the “Fixed_Outlet” and “Time_Modulated”
options.
3.7. FIXED-OUTLET PRV

The fixed outlet PRV option is usually the first option to be considered since it is the
easiest option to implement and also the least expensive. The objective is to lower the
zone inlet pressure in order to lower the pressure in the system to the level where the
minimum acceptable pressure occurs during the period of maximum system demand.
This option can only be considered in cases where the pressure during the period of
maximum demand exceeds the minimum level of service. In many systems it is found
that the pressure during the period of maximum demand is below the acceptable fire-
fighting requirements or other minimum pressure as defined in the local by-laws. In such
cases there is no scope for further pressure reduction using a fixed outlet PRV, although
there may be some scope for improvement using the time-modulated option.
The data screen for the fixed-outlet PRV option is shown in Fig. 3.9.
Figure 3.9: Typical data entry form for the Fixed-Outlet PRV Option
3—12 02/05/16
It should be noted that the only user input is the pressure setting for the proposed Fixed-
Outlet PRV which is indicated by the red value of 50 m in Fig. 3.9.
3.8. TIME-MODULATED PRV

The time-modulated PRV option involves a standard fixed-outlet PRV which is equipped
with a time-modulating controller. The maximum pressure setting is set on the main PRV
while the low setting is set on the electronic controller. After the high and low pressure
settings have been defined, the switching times must then also be derived. Either two
intervals or four intervals can be selected and the model caters for both. The typical input
form for the time-modulated PRV option is shown in Fig. 3.10.
Fig. 3.10: Example of the Time -Modulated PRV Form
On Fig. 3.10 it can be seen that the user has decided to try maximum and minimum
pressures of 50 m and 30 m respectively and that two periods of high pressure have been
selected: the first from 6 am to 8 am and the second from 5 pm to 8 pm. In addition, it can
also be seen that in this case the minimum acceptable pressure of 19 m (defined in the
data_input form under general_info) is not achieved during all 24 h in the simulation as
noted in the “status” box. In a case like this the user would then adjust the pressures or
the switching periods (or both) until the maximum savings are achieved without the
pressure dropping below the minimum acceptable value.
3—13 02/05/16
3.9. SUMMARY
The summary form provides a very simple and concise table of the costs associated with
the different options compared to the savings achieved for the different options
considered. In order to obtain a sensible summary, the user must supply certain
information including:
• The costs associated with the options considered;
• The cost of inflow (R/m 3);
• Consumption value (R/m 3);
• A multiplication factor (%);
• % of residential consumption judged to be pressure-dependent;
• % of non-residential consumption judged to be pressure-dependent;
• N1 value for correction to pressure-dependent consumption.
The layout of the summary form is shown in Fig. 3.11.
Figure 3.11: Example of the Summary Form
3—14 02/05/16
Costs associated with development options

The costs associated with the development options are supplied by the user on a
separate form which is accessed by clicking on the “Implementation Costs” button which
is on the main Summary form. This brings up a new form where the user can define the
items to be used and the appropriate cost as shown in Fig. 3.12.
Figure 3.12: Example of “Implementation Costs” Form
The cost of inflow (R/m3) and Consumption value (R/m3)

The cost of inflow is the cost of the water entering the zone to the supplier. Any savings in
leakage reduction before the meter will result in a saving to the supplier. Any reduction
in water use by the consumer after the meter will represent a saving on water purchases
but will also represent a loss of revenue to the water supplier at the average selling price.
Any reductions in use or leakage after the meter will therefore represent a loss of revenue
to the water supplier at a rate equal to the difference in the buying and selling price of the
water.
Percentage of water use judged to be pressure-dependent and N1 value

As mentioned previously, if pressure is reduced there is normally some reduction in water
use by the consumer. In order to take the potential loss of revenue into account, it is
3—15 02/05/16
necessary to make some estimate of the percentage of use that is thought to be pressure-
dependent. This is, unfortunately, rather difficult to establish and the user must make an
estimate based on experience of the system in question. In the example shown in
Fig. 3.11, values of 10% and 2% were selected for the pressure-dependent residential
and non-residential use respectively. In addition, a separate N1 value is provided to
assess the likely reduction in use due to the reduced pressure. The N1 value used for the
overall system leakage is not appropriate since most of the reduction in consumer use will
occur through a tap or other orifice where an N1 value of 0.5 is more likely to be
appropriate.
Multiplication Factor (%)

A multiplication factor has been provided to allow the user to scale down the predicted
savings to avoid over-optimistic predictions. In some cases it may be considered
appropriate to leave this factor at 100% since the results are already conservative to some
degree since they do not include the savings through reduced bursts and the associated
repair costs etc. In some instances, however, the user may feel more comfortable which
scaling down the predicted savings since the 24-hour period upon which the analysis is
based is perhaps a particular sever period and not representative of the year as a whole.
In such cases the user can change the factor in accordance with the particular
circumstances of the zone in question to take account of seasonal demand patterns etc.
3—16 02/05/16
4. REFERENCES
LAMBERT, A.O.,1997 Pressure Management/Leakage Relationships: Theory,

Concepts and Practical Applications. Conference on Minimising Losses in Water
Supply Systems, Apr. 1997. IQPC Ltd, London.
LAMBERT, A, MYERS, S and TROW,S. (1998) Managing Water Leakage :

Economic and Technical Issues. Financial Times Energy Publications, Maple
House, 149 Tottenham Court Road, London W1P 9LL, ISBN 1 84083 011 5
LAMBERT A., BROWN T.G., TAKIZAWA M., WEIMER D, 1999. A Review of

Performance Indicators for Real Losses from Water Supply Systems. AQUA,
Dec 1999. ISSN 0003-7214
MAY, J. 1994. Pressure Dependent Leakage. World Water and Environmental

Engineering, October 1994
NATIONAL WATER COUNCIL STANDING TECHNICAL COMMITTEE,1980.

Report No. 26, Technical Working Group on Waste of Water. Leakage Control
Policy and Practice.. ISBN 0 904561 95 X
OFFICE OF WATER SERVICES, 1998. 1997/98 Report on leakage and water

efficiency. OFWAT, ISBN 1 874234 42 6
UK WATER INDUSTRY, 1994. The ‘Managing Leakage’ Series of Reports
Report A : Summary Report. ISBN: 1 898920 06 0
Report B : Reporting Comparative Leakage Performance. ISBN: 1 898920 07 9
Report C : Setting Economic Leakage Targets. ISBN: 1 898920 08 7
Report D : Estimating Unmeasured Water Delivered. ISBN: 1 898920 09 5
Report E : Interpreting Measured Night Flows. ISBN: 1 898920 10 9
4—1 02/05/16
Report F : Using Night Flow Data. ISBN: 1 898920 11 7
Report G: Managing Water Pressure. ISBN: 1 898920 12 5
Report H: Dealing with Customers’ Leakage. ISBN: 1 898920 13 3
Report J: Leakage Management Techniques, Technology & Training.
ISBN: 1 898920 14 1
WRC, 1999. Development of a standardised approach to burst and

background losses in water distribution systems in South Africa. South
African Water Research Commission, Report No. TT 109/99, June 1999,
ISBN 1 86845 490 8
4—2 02/05/16
APPENDIX A
Introduction To Babe And Favad

Concepts, And Calculation Of
Unavoidable Annual Real Losses
Appendix A 02/05/16
APPENDIX A : INTRODUCTION TO BABE AND FAVAD CONCEPTS, AND

CALCULATION OF UNAVOIDABLE ANNUAL REAL LOSSES
A1: HISTORICAL BACKGROUND

As a result of the privatisation of the England & Wales Water Service Companies in
1989, it became necessary for all water suppliers to be able to demonstrate to their
regulators that they fully understood their position on leakage. This did not imply
that all water suppliers had to achieve the lowest possible leakage levels, but simply
that correct and appropriate technical and economic principles were being applied to
leakage management.
Accordingly, in 1990 a National Leakage Control Initiative (NLCI) was established in

England & Wales by the Water Services Association and the Water Companies
Association, to update and review the ‘Report 26’ guidelines (National Water
Council, 1980) for leakage control that had been in use in the UK since 1980.
Considerable progress had been made in equipment and metering technology over
the previous ten-year period, but methods of data analysis had not kept pace with
these technical improvements.
In order to co-ordinate the various research efforts, described in the ‘Managing

Leakage’ series of Reports (UK Water Industry, 1994), Mr Allan Lambert, then
Technical Secretary of the NLCI, developed an overview concept of components of
real losses, and the parameters which influence them. This concept, based on
internationally applicable principles, is known as the Burst and Background
Estimates (BABE) methodology. The BABE concepts were first applied and
calibrated in the UK, and three simple pieces of standard software using the BABE
concepts were made available at the time of issue, in 1994, of the ‘Managing
Leakage’ Reports.
Prior to 1994, a single relationship between minimum night flow and pressure was
normally assumed in the UK, based on the ‘Leakage Index’ curve in Report 26. The
1994 ‘Managing Pressure’ Report recognised that there was not a single
relationship, but did not offer an alternative method. However, a much improved
understanding of the range of relationships between pressure and leakage rate was
introduced separately from the ‘Managing Leakage’ Reports in 1994, when John
May published his FAVAD (Fixed and Variable Areas Discharges) concept
(May, 1994). Using FAVAD, it has been possible to reconcile apparently diverse
Appendix A A- 1 02/05/16
relationships and data from laboratory tests and distribution sector tests in Japan,
UK, Brazil, Saudi Arabia, and Malaysia,
Since 1994, the BABE and FAVAD concepts have been applied in many countries
for the solution of a wide range of leakage management problems, as described by
Lambert (1997).
Fig. A.1 shows the typical range of problems that can be successfully tackled with
these concepts. The remainder of this Appendix explains the application of BABE
and FAVAD concepts to the development of the International Performance
Indicators for real losses.
Figure A.1: Problem-Solving using BABE and FAVAD concepts
A2: BURST AND BACKGROUND ESTIMATE (BABE) procedures

In order to address leakage, it was considered necessary to first understand the
various components making up the water balance for a typical water supply network.
The previous approach as shown in Fig. A.2 was to consider three main
components: Authorised Metered, Authorised Unmetered and the remainder which
represents all unaccounted-for water, and is often referred to as the real and
apparent losses. Further details on real and apparent losses are provided later in
this section and are also shown in Fig. A.4.
Real and
Apparent Losses
Authorised
Unmetered
Delivered
Authorised
Metered
Delivered
Figure A.2: Traditional Water Balance.
In view of the large portion of the traditional water balance that was usually
represented by the real and apparent losses, the whole water balance approach
was revised by breaking the balance down into smaller components that could either
be measured or estimated. In this manner it was possible to gain a greater
understanding of the different components and also of their significance to the
overall water balance. A typical example of the BABE water balance is provided in
Fig. A.3. It should be noted that the water balance need not be restricted to the
components shown in this figure and, conversely, it can be split into a greater
number of components or perhaps different components. Every system is different
and it is the general approach that should be applied and not a specific and rigid
framework.
Figure A.3: BABE Water Balance Approach.
The BABE water balance approach has now been widely accepted worldwide and is
also incorporated in much of the latest South African water legislation. It is not a
highly technical or complicated approach. On the contrary, it is extremely simple
and logical. The typical components that can be included in any particular water
balance were established at the International Water Supply Association Workshop
held in Lisbon in May 1997. The water balance components identified at the
workshop are shown in Fig. A.4. It should be noted that the components shown in
this figure also include the losses associated with the bulk water system as well as
the purification system. For municipalities supplying only the water on the
distribution side of the bulk supply system, many of the items shown in Fig. A.4 can
be omitted. Similarly, in many of the municipalities in South Africa, the internal
plumbing losses (LP) dominate the whole water balance, although such losses are
represented by only a small block in the figure. In such cases it may not be
necessary to undertake a full and detailed water balance until the plumbing losses
are under control.
Figure A.4: Recommended BABE Water Balance Components.
Fig. A.4 provides a breakdown of the most important components that can be
included in a water balance for a specific water supplier. It is important to note that
the losses have been broken down into real and apparent losses. Real losses are
those where the water has, in fact, left the system and has not been utilised in any
way. If such losses can be reduced, the total water required by the supplier will also
be reduced. Apparent losses, on the other hand, are simply “paper” losses that do
not represent a loss from the system. They are usually due to illegal connections,
and meter and billing errors. If such losses are eliminated, the total water required
by the supplier may not change. However, the “unaccounted-for” component in the
water balance will be reduced. In such cases certain other components such as
“Authorised Metered” or even “Authorised Unmetered” will increase as the apparent
losses are reduced.
A3: WHAT ARE BURST AND BACKGROUND LEAKS ?

The larger detectable events are referred to as bursts, while those that are too small
to be located (if not visible) are referred to as background leaks. The threshold
between bursts and background leaks can vary from country to country, depending
upon factors such as minimum depth of pipes, type of ground and surface, etc. In
the UK a threshold limit of 500 litres/hour was used in the 1994 Managing Leakage
Reports, but advances in technology and other factors suggest that a figure of
around 250 litres/hour would be more appropriate in South Africa. In other words:
Events > 250 litres/hour = Bursts
Events < 250 litres/hour = Background Leaks
In all water supply systems there are likely to be both bursts and background leaks
since it is not possible to develop a system completely free from leakage. However,
using the BABE concepts, it is possible to calculate the unavoidable annual real
losses on a system-specific basis.
A4: USE OF FAVAD AND BABE CONCEPTS IN DEVELOPMENT OF

PERFORMANCE INDICATORS
The best of the traditional; basic (IWA Level 1) Performance Indicator for
Operational management of real losses is the following:
Litres/service connection/day (when the system is pressurised)
This basic Operational Performance Indicator, however, does not take account of
three system-specific key factors which can have a strong influence on lowest
volume of Real Losses which can be achieved in any particular system. These are:
• Average operating pressure;

• Location of customer meters on service connections (relative to the
street/property boundary);
• Density of service connections (per km of mains).
The ‘Intermediate’ Operational Performance Indicator for Real Losses, deals

with the first of these key factors by assuming a linear relationship between average
leakage rate and pressure, i.e. the Intermediate Performance Indicator becomes:
Litres/connection/day/metre of pressure (when the system is pressurised)
The justification for this assumption can be explained using the FAVAD concept. In
its simplest form, this assumes that leakage rate (L) varies with Pressure (P) to the
power N1, i.e.
N1
L varies with P
International research has shown that different types of leakage paths have different
values of N1, which can range from 0.5 to 2.5. Values of N1 derived from tests on
small sectors of distribution systems are usually in the range 0.5 to 1.5. When a
weighted average of these N1 values is calculated for application to larger
distribution systems, the average N1 value is usually quite close to 1.0 (Lambert,
1997), i.e a linear relationship can be assumed.
The ‘Intermediate’ Operational Performance Indicator does not, however, deal with
the second and third of the system-specific key factors which can influence the
lowest volume of real losses which can be achieved in any particular system, i.e.
• Location of customer meters on service connections (relative to street/property

boundary);
• Density of service connections (per km of mains).
The ‘Detailed’ Operational Performance Indicators for Real Losses, deals with
both these factors, and average operating pressure, by calculating a system-specific
value for ‘Unavoidable Annual Real Losses’ (UARL). The ratio of the Current Annual
Real Losses (CARL, calculated from the standard Water Balance) to the UARL, is
the Infrastructure Leakage Index (ILI), i.e.
Infrastructure Leakage Index ILI = CARL/UARL
The equation for UARL is based on BABE (Background and Bursts Estimates)
concepts. With the BABE concepts, it is possible to calculate, from first principles,
the components which make up the annual volume of Real Losses. This is because
the leaks occurring in any water supply system can be considered conceptually in
three categories:
• Background leakage – small undetectable leaks at joints and fittings;
• Reported bursts – events with larger flows which cause problems and are
reported to the water supplier;
• Unreported bursts – significant events that do not cause problems and can only
be found by active leakage control.
A5: CALCULATION OF UNAVOIDABLE ANNUAL REAL LOSSES (UARL)

The procedure to estimate the unavoidable annual real losses (UARL) was
developed by Lambert during the period of the International Water Association’s
Task Force on Water Losses. The methodology is described in a paper in AQUA
(Lambert et. al., 1999) and basically involves estimating the unavoidable losses for
three components of infrastructure, namely:
• Transmission and distribution mains (excluding service connections);

• Service connections, mains to street/property boundary;
• Private underground pipe between street/property boundary and customer meter.
In South Africa, the third of these components can normally be ignored since
customer meters are located close to the edge of the street.
The parameters used in the calculation of the losses are indicated in Table A1.
From this table it can be seen that the one variable common to all elements is
pressure. This is also the one variable that is normally excluded from most
commonly used leakage performance indicators such as percentage, leakage per
connection per year and leakage per km of mains per year, etc.
Each of the elements in Table A1 can be allocated a value appropriate to

infrastructure in good condition, operated in accordance with best practice, based on
the analysis of data from numerous systems throughout the world. The results are
provided in Table A2.
The parameter values indicated in Table A2 include data for minimum background
loss rates and typical burst frequencies for infrastructure in good condition, and for
typical average flow rates of bursts and background leakage at 50m pressure. The
average duration assumed for reported bursts is based on best practice world-wide.
The average duration for unreported bursts is based on intensive active leakage
control, approximating to night flow measurements once per month on highly
sectorised water distribution systems.
Table A1: Parameters required for calculation of UARL
Component of Background Reported Unreported

Infrastructure Losses Bursts bursts
Mains • Length • Number/year • Number/year

• Pressure • Pressure • Pressure
• Minimum loss rate/km* • Average flow rate* • Average flow rate
• Average duration • Average duration
Service • Number • Number/year • Number/year
connections to
• Pressure • Pressure • Pressure
street/property line
• Minimum loss rate/conn* • Average flow rate* • Average flow rate
Service • Length • Number/year • Number/year
connections after • Pressure • Pressure • Pressure
• Minimum loss rate/km* • Average flow rate* • Average flow rate
* these flow rates are initially specified at 50m pressure
Table A2: Parameter values used to calculate UARL
Component of Background Reported Unreported

Infrastructure Losses Bursts Bursts
Mains 20* • 0.124 bursts /km.yr at • 0.006 bursts /km.yr at
3
l /km.h • 12 m /h per burst* • 6 m 3/h per burst*
• average duration of 3 • average duration of 50 days
days
Service connections 1.25* • 2.25/1000 • 0.75/1000 conn.yr at
to street/property connections.yr at
l /conn.h • 1.6 m 3/h per burst*
line 3
• 1.6 m /h per burst* • average duration of 100 days
• average duration of 8
days
Unmetered Service 0.50* • 1.5/1000 connections.yr • 0.50/1000 conn.yr at
connections after at
l /conn.h • 1.6 m 3/h per burst*
street/property line 3
• 1.6 m /h per burst*
per 15m length • average duration of 101 days
• average duration of 9
days
* these flow rates are initially specified at 50m pressure
Assuming a simplified linear relationship between leakage rate and pressure, the
components of UARL can be expressed in modular form, for ease of calculation, as
shown in Table A3. Sensitivity testing shows that differences in assumptions for
parameters used in the ‘Bursts’ components have relatively little influence on the
‘Total UARL’ values in the 5th column of Table A3.
Table A3: Calculated Components of Unavoidable Annual Real Losses (UARL)
Component of Background Reported Unreporte Total Units

Infrastructure Losses Bursts d UARL
Bursts
Mains 9.6 5.8 2.6 18 l /km mains/day
per m of pressure
Service 0.60 .04 0.16 0.8 l /connection/day/
connections to m of pressure
Unmetered Service 16.0 1.9 7.1 25 l /km underground
connections after pipe/day/metre of
street/property line pressure
NOTE: the UARL losses from Unmetered Service Connections after the street/property line can be
ignored in the South African context, as all customers are metered and these meters are located close
to the street/property line. The losses from the service connections (main to meter) tend to dominate
the calculation of UARL in most parts of South Africa, except at low density of connections (less than
20 per km of mains).
Based on the figures provided in Table A3, the calculation of the UARL can be
expressed as follows:
UARL = (18 * Lm + 0.80 * Nc + 25 * Lp) * P
Where:
UARL = Unavoidable annual real losses (l/day)
Lm = Length of mains (km)
Nc = Number of service connections (main to meter)
Lp = Length of unmetered underground pipe from street edge to
customer meters (km)
P = Average operating pressure at average zone point (metres)
Example: A system has 114 km of mains, 3 920 service connections all located at
the street property boundary edge and an average operating pressure of 50 m.
UARL = (18 * 114 + 0.80 * 3920 + 25 * 0) * 50 l/day

= 102 600 + 156 800 l/day
= 259 400 l/day
= 259.4 m 3/day
= 94 681 m 3/year
Appendix A A - 10 02/05/16
APPENDIX B
Data Capture Form for PRESMAC
Appendix B 02/05/16
Appendix B B-1 02/05/16

(WRC-SouthAfrica-2001-R. McKenzei) PRESMAC User Guide

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(WRC-SouthAfrica-2001-R. McKenzei) PRESMAC User Guide

Uploaded by

Copyright:

Available Formats

Pressure Management Program (PRESMAC) TT152/01

SOUTH AFRICAN WATER RESEARCH

Ronnie McKenzie : WRP Pty Ltd.

(Coding by Stephen Langenhoven)

WRC Report No TT 152/01

Water Research Commission

ISBN No. 1 86845 722 2

Pressure Management Program (PRESMAC)

The PRESMAC Model

Although the pressure management software developed in the UK is available

• expected leakage rates from connections, properties and mains;

• pressure exponent for the system as a whole;

• details of any commercial consumers.

• pressure at the inlet point;

• pressure at the average zone point;

• pressure at the critical point;

• inflow to the zone.

1. PRESSURE MANAGEMENT MODEL ............................................................................1—1

1.1. GENERAL CONCEPTS OF PRESSURE MANAGE MENT....................................................1—1

1.2. CONCEPTS OF ACTIVE PRESSURE CONTROL................................................................1—3

1.3. THE PRESMAC MODEL.......................................................................................................1—10

1.4. METHODOLOGY ..................................................................................................................1—17

1.5. PURPOSE AND LAYOUT OF THIS REPORT.....................................................................1—18

2. DETAILS OF THE PRESMAC MODEL...........................................................................2—1

2.1. HOW THE MODEL W ORKS...................................................................................................2—1

2.2. PRESSURE EXPONENT (N1)................................................................................................2—2

2.3. CALCULATION OF THE HOUR-DAY FACTOR ....................................................................2—4

2.4. PRESSURE-DEPENDENT AND PRESSURE-INDEPENDENT FLOWS..............................2—5

2.5. ESTIMATION OF NETWORK FRICTION (“K”) FACTORS....................................................2—9

2.6. ANALYSIS OF FIXED-OUTLET PRV...................................................................................2—11

2.7. ANALYSIS OF TIME-MODULATED CONTROLLER...........................................................2—18

3. USING PRESMAC ..............................................................................................................3—1

3.1. HARDWARE AND SOFTWARE REQUIREMENTS...............................................................3—1

3.2. INSTALLING PRESMAC.........................................................................................................3—1

3.3. DATA REQUIREMENTS.........................................................................................................3—1

3.4. RUNNING PRESMAC .............................................................................................................3—2

3.5. DATA INPUT ...........................................................................................................................3—3

3.6. CURRENT SITUATION...........................................................................................................3—9

3.7. FIXED-OUTLET PRV............................................................................................................3—12

3.8. TIME-MODULATED PRV......................................................................................................3—13

3.9. SUMMARY ............................................................................................................................3—14

Figure 3.6 Tabulated Information from the Current_Situation Form

APPENDIX A: Introduction to BABE and FAVAD Concepts, and Calculation

APPENDIX B: Data Capture Form for PRESMAC

1. PRESSURE MANAGEMENT MODEL

1.1. GENERAL CONCEPTS OF PRESSURE MANAGEMENT

Most water distribution systems experience significant fluctuations in demand throughout

1.2. CONCEPTS OF ACTIVE PRESSURE CONTROL

• Fixed outlet PRV;

Day 1 Day 2 Day 3 Day 1 Day 2 Day 3

Figure 1.3: Pressure control using conventional PRV

Figure 1.4: Typical 4-point time modulated pressure profile

Figure 1.5: Typical installation of a time -modulated PRV controller

Day 1 Day 2 Day 3 Day 1 Day 2 Day 3

Time Modulated time time

Inlet Outlet Extremity

Figure 1.6: Pressure control using a time -modulated PRV controller

Photograph courtesy of Technolog, UK.

Figure 1.7 : Components required for a flow-modulated PRV installation

Flow Modulated Day 1 Day 2 Day 3 Day 1 Day 2 Day 3

Controller time time

Inlet Outlet Extremity

Figure 1.8: Pressure control using a flow-modulated PRV controller

Pressure Critical node